Description of the CMAQ-urban model
Sean D Beevers, Nutthida Kitwiroon, Martin L Williams, David C Carslaw
MRC-HPA Centre for Environment and Health, Kings’ College London, United Kingdom
1 Method
1.1 Introduction
The Weather Research and Forecasting (WRF) meteorological model1, the Community Multiscale Air
Quality (CMAQ) regional scale model2 and the Atmospheric Dispersions Modelling System (ADMS) roads
model 3 have been coupled to create CMAQ-urban. WRF provides the meteorological input to the CMAQ
model which provides air quality predictions starting at a European scale and through a series of model
nests which reduce in size, finally focus on Greater London. The CMAQ model provides hourly estimates
of air quality concentrations across a 3 x 3 km London grid, and these outputs are coupled with the
ADMS model, also driven using WRF meteorology, to provide fine scale hourly air quality predictions .
The CMAQ-urban model gives hourly average concentrations of NOX, NO, NO2, O3 and PM components
at 20m x 20m resolution in London. The setup of each component, described in sections 1.2 to 1.5, is
taken from Beevers et al4.
1.2 WRF setup
The WRF model v3.1 was set up using 23 vertical layers with 7 layers under 800m and the top layer
~15km above the ground. Four horizontal nesting levels were used, downscaling from 81km grids over
Europe to 27km over the UK and Ireland, 9km over the UK and 3km over London. Initial and boundary
(IC/BC) conditions were taken from the GFS model at 1x1 degree
(http://www.esrl.noaa.gov/psd/data/reanalysis/reanalysis.shtml ). Other physics settings within WRF
included use of the Rapid Radiation Transfer Model (RRTM) radiation scheme5, the Kain-Fritsch (new
Eta) microphysics scheme6, the Yonsei University (YSU) planetary boundary layer scheme7, the MoninObukhov surface scheme1 and the National Ocean and Atmospheric Administration (NOAA) land surface
scheme8. Observational nudging of wind speed, temperature and relative humidity was used in the WRF
model with a nudging strength of 6x10-4 and radius of influence of 200 km. In urban areas urban land
cover and Z0 values of 0.8m were used. No anthropogenic heat from buildings or road transport was
included in the WRF model.
1.3 CMAQ and emissions inventory setup
The CMAQ v4.6 model was set up using the same vertical and horizontal grid structure as for WRF.
Atmospheric chemistry was simulated using the CB 05 chemical mechanism9 and initial and boundary
conditions, included as monthly means, from the UK Meteorological Office chemistry-transport model,
1
STOCHEM (Derwent pers comm.), which provided 14 pollutant species including CO, NH3, NOX, O3, HNO3
and VOC’s. Anthropogenic and natural emissions were derived from the European Monitoring and
Evaluation Programme (EMEP, http://www.ceip.at/), the European Pollutant Release and Transfer
Register (EPRTR), the National Atmospheric Emissions Inventory (NAEI)10 and the London Atmospheric
Emissions Inventory (LAEI)11. The emissions were summarised into 11 Selected Nomenclature for
sources of Air Pollution or SNAP sectors and were processed into gridded chemical species using
speciation and temporal profiles developed in the US-EU project, Air Quality Modelling Evaluation
International Initiative (AQMEII) (http://aqmeii.jrc.ec.europa.eu/). Hourly emissions for road traffic were
created using a time series of normalised traffic counts, taken from 37 automatic sites in London. The
monthly and hour of week profiles for each SNAP sector were processed using the Sparse Matrix
Operator Kernel Emissions model (SMOKE v 2.4). Biogenic emissions were estimated using high
resolution Coordination of Information on the Environment (CORINE) land cover data
(http://www.eea.europa.eu/data-and-maps), meteorological data from WRF and methods described by
Guenther et al12 and Sanderson13.
The input emissions data was provided at a range of spatial scales, from 50km grids across Europe
(EMEP) down to 1 km grids for the NAEI and LAEI. The link between each emissions source and the
CMAQ model domains was undertaken by re-projecting all of the emissions data into the Lambert
Conformal projection and then intersecting each emissions database and model domain in ArcGIS v9.3.
Where a CMAQ grid cut an emissions grid into two or more pieces, a proportion of the emissions was
added to the model grid, based upon the intersected area.
Road transport NO and exhaust primary NO2 emissions for the UK and London, were taken from the
NAEI and the LAEI, respectively, and released into the lowest layer of the model, 14m high. Assumptions
for exhaust primary NO2 emissions were taken from Grice et al14. Large industrial plant emissions from
~10,000 sources in the EU were taken from the NAEI and EPRTR databases and included major power
stations, iron and steel smelting and oil refineries. For 100 UK sources, details of stack heights , stack
diameters, volume flow rates and release temperatures were used by SMOKE to account for the effects
of plume rise. In London, commercial and domestic gas combustion for space heating, which represents
a large source of NOX11, and all other ground based sources of NOX were released into the lowest 14m
layer of the CMAQ model.
1.4 ADMS setup
The ADMS roads model (v2.3)3 was used to describe the near field dispersion from roadways in CMAQurban, using the hourly meteorological inputs: wind speed and direction, temperature, surface sensible
heat flux and planetary boundary layer height, predicted from the WRF model. The ADMS model was
run for each hour of the year using similar methods to those described in Kelly et al15, to produce hourly
concentration fields or model kernels. Each kernel represents the concentration of a primary pollutant
(with no chemistry applied) across a regular grid as it dilutes away from a road source of unit emissions.
The concentrations across each kernel were predicted at 5m intervals and within 225m of each source,
2
using a constant road emissions rate of 1 g km-1 s-1. Road geography is highly important and so the
emissions from roads in London were represented using the centreline of each carriageway divided into
10m sections. The 10m granularity of the road sources was considered to be representative of complex
road curves but also the distance between carriageways on larger roads such as motorways. Six road
categories (and associated kernels) were used in London, including open roads (motorway), typical
roads (average urban roads surrounded by low rise buildings) and 4 types of street canyon (classified by
their orientation: north-south, east-west, southwest-northeast and southeast-northwest). The “typical
roads” had a road width of 20m and a building height of 10m and “street canyons” had a width of 30m
and a building height of 25m. The “street canyon” width and height details were manually sampled from
the 3D building model in London (http://www.casa.ucl.ac.uk/news/newsStory.asp?ID=80). Once
created, each kernel was applied to the 1746 major road emissions estimates in the LAEI consisting of
~340,000 10m road sources, and the hourly concentration of NO and primary exhaust NO2 combined
onto a fixed grid of 20m x 20m in London.
The near road chemistry was simulated using a simple chemical scheme described in Carslaw and
Beevers16. The reaction rates and photo-dissociation rates were taken from the photolysis rate preprocessor (JPROC)17, part of the CMAQ run, and the time of flight from road sources, estimated each
hour, as a concentration weighted average at each receptor location, assuming a straight line between
source and receptor and using WRF wind speed at 10m height.
1.5 CMAQ-urban outputs
The CMAQ predictions of hourly NO, NO2 and O3 at 3 x 3 km in London were downscaled to a fixed
20mx20m grid using the bilinear interpolation method described in Shao18. The CMAQ hourly
concentrations are smoothly varying in space for each hour and this limits the error introduced through
the interpolation step and allows a direct match with the ADMS model outputs. The ADMS predicted
concentrations of NO and primary exhaust NO2 were then added to the CMAQ NO, NO2 and O3
concentrations, over the fixed 20m grid, and the simple chemistry scheme applied to give 87.2 billion
hourly predictions of NO, NO2 and O3 across London.
1.6 REFERENCES
1.Skamrock, W.C., J.B. Klemp, J. Dudhia, D.O. Gill, D.M. Barker, M.G. Duda, X- Y. Huang, W. Wang, and J.G. Powers.
2008: A Description of the Advanced Research WRF Version 3. NCAR/TN–475+STR.
2. Byun, D.W. and J.K.S. Ching. 1999: Science Algorithms of the EPA Models-3 Community Multiscale Air Quality
(CMAQ) Modeling System. U.S. Environmental Protection Agency, Office of Research and Development.
EPA/600/R-99/030.
3. CERC. 2006: ADMS-Roads An Air Quality Management System User Guide. Cambridge Environmental Research
Consultants Ltd, Cambridge. (http://www.cerc.co.uk/environmentalsoftware/assets/data/doc_userguides/CERC_ADMS-Roads%20Extra3.1_User_Guide.pdf). Accessed
03/01/2012
4. Beevers SD, Kitwiroon N, Williams ML, Carslaw DC. 2012. One way coupling of CMAQ and a road source
dispersion model for fine scale air pollution predictions in London. Atmospheric Environment 59, pp 4758.
3
5. Mlawer, E. J., S. J. Taubman, P. D. Brown, M. J. Iacono, and S. A. Clough, 1997: Radiative transfer for
inhomogeneous atmosphere: RRTM, a validated correlated-k model for the longwave. J. Geophys. Res.,
102 (D14), 16663–16682.
6. Kain, J. S., 2004: The Kain-Fritsch convective parameterization: An update. J. Appl. Meteor., 43, 170–181.
7. Hong, S.-Y., Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of
entrainment processes. Mon. Wea. Rev., 134, 2318–2341.
8. Chen, F., and J. Dudhia, 2001: Coupling an advanced land-surface/ hydrology model with the Penn State/ NCAR
MM5 modeling system. Part I: Model description and implementation. Mon. Wea. Rev., 129, 569–585.
9. Yarwood, G., Rao, S.T., Yocke, M., Whitten, G., 2005. Updates to the Carbon Bond Chemical
476 mechanism: CB05, report, Rpt. RT-0400675, US EPA, Res. Tri. Park.
10. Murrells TP, Passant NR, Thistlethwaite G, Wagner A, Li Y, Bush T, Norris J, Walker C, Stewart RA, Tsagatakis I,
Whiting R, Conolly C, Okamura S, Peirce M, Sneddon S, Webb J, Thomas J, MacCarthy J, Choudrie S,
Brophy N. 2010. UK Emissions of Air Pollutants 1970 to 2008. Report produced for Department for
Environment,
Food
and
Rural
Affairs
(DEFRA).
http://ukair.defra.gov.uk/reports/cat07/1009030925_2008_Report_final270805.pdf. Accessed 31/05/2011.
11. Greater London Authority (GLA). 2010. The London Atmospheric Emissions Inventory 2008.
http://data.london.gov.uk/laei-2008 Accessed 31/05/2011.
12. Guenther, A., C.N. Hewitt, D. Erickson, R. Fall, C. Geron, T. Graedel, P. Harley, L. Klinger, M. Lerdau, W.A.
McKay, T. Pierce, B. Scholes, R. Steinbrecher, R. Tallamraju, J. Taylor and P. Zimmerman, 1995. A global
model of natural volatile organic compound emissions. Journal of Geophysical Research, 100, 8873–8892.
13. Sanderson, M.G. 2002. Emission of isoprene, monoterpenes, ethene and propene by vegetation, Hadley Centre
Technical Note 40, UK Meteorological Office.
14. Grice, S., Stedman, J., Kent, A., Hobson, M., Norris, J., Abbott, J., Cooke, S., 2009. Recent trends and projections
of primary NO2 emissions in Europe. Atmospheric Environment 43 (13), 2154 – 2167.
http://www.sciencedirect.com/science/article/ B6VH3-4VDS8MT3/2/4a692307ac35ebf6ba37560cf32b4cf9
15. Kelly FJ, Anderson HR, Armstrong B, Atkinson RW, Barratt B, Beevers SD, Mudway IS, Green D, Derwent RG,
Tonne C, Wilkinson P. 2011. The impact of the Congestion Charging Scheme on air quality in London.
Health Effects Institute. http://pubs.healtheffects.org/view.php?id=358. Accessed 16/07/11.
16. Carslaw DC. and Ropkins K. 2012. Openair — an R package for air quality data analysis. Environmental Modelling &
Software. Volume 27-28, 52-61.
17. Roselle., S.J., Schere, K.L. and Pleim, J.E. 1999. Chapter 14: Photolysis rates for CMAQ, In: Science Algorithms of
the EPA Models-3 Community Multiscale Air Quality (CMAQ) Modeling System. EPA/600/R-99/030
18. Shao, X. Stein, M. and Ching, J. 2007. Statistical comparisons of methods for interpolating the output of a
numerical air quality model, Journal of Statistical Planning and Inference, 137, 2277 – 2293.
1.7 ACHKNOWLEDGEMENTS
The authors would like to acknowledge USEPA and NCEP/NCAP for CMAQ/WRF models, Department of
Environment, Food and Rural Affairs (DEFRA) for the UK emissions and monitoring data and King’s
College for road traffic emissions and monitoring data in London. Also thanks go to the AQMEII project
for use of their hourly emissions profiles. We also acknowledge the Natural Environment Research
Council, Medical Research Council, Economic and Social Research Council, Department of Environment,
Food and Rural Affairs and Department of Health for the funding received for the Traffic Pollution and
Health in London project (NE/I008039/1), funded through the Environmental Exposures & Health
Initiative (EEHI).
4
Description of the KCLurban model
Sean D Beevers, Mark De Jong, David Dajnak, David C Carslaw
MRC-HPA Centre for Environment and Health, Kings’ College London, United Kingdom
2 Methods
2.1 Model summary
In the KCLurban model used was made of a kernel modelling technique, based upon the ADMS1,2, to
describe the initial dispersion from each emissions source. The contribution from each source was
summed onto a fixed 20m x 20m grid across London assuming that one can calculate the contribution of
any source to total air pollution concentrations by applying each kernel and adjusting for the source
strength. The kernels were produced using an emissions source of unity, either 1 g s-1 (point and jet
sources), 1 g m-3 s-1 (volume sources) or a 1 g km-1 s-1 (road sources) and were created using hourly
meteorological measurements from the UK Meteorological Office site at Heathrow. Data from the
Heathrow site is recorded at a height of 10 metres and includes measurements of temperature, wind
speed, wind direction, precipitation, relative humidity and cloud cover.
The KCLurban model is computationally efficient and provides annual mean NOX, NO2 , O3, PM10 and
PM2.5 at a resolution of 20m x 20m. Predictions in London were based upon the London Atmospheric
Emissions Inventory (LAEI)3.
2.2 Treatment of sources in KCLurban
To predict air quality concentrations in London the KCLurban model sums together three source
categories. First, for rural sources outside the model domain, we used rural NOx measurements and for
PM10 and PM2.5 we used the well established modelling techniques of Fuller4 and Fuller and Green5,6.
Second, within the model domain, but greater than 500m from a receptor location, we represented
sources as shallow volumes of 1x1 km horizontal dimension and 2m vertically for road traffic, and 50m
vertically for other sources. Third, for those sources within 500m of a receptor location, we used a
detailed treatment of road/rail/aircraft and gas combustion sources each of which is described in
sections 2.3 to 2.6.
Sources in the KCLurban model include: road transport (exhaust and non-exhaust), large regulated
industrial processes, small regulated industrial processes, large boiler plant, gas heating (domestic and
industrial-commercial), oil combustion sources (domestic and commercial), coal combustion sources
(domestic and commercial), agricultural and natural sources, rail, ships, airports, others (sewage plant
5
etc) and biomass burning. In modelling the emissions from large industrial processes use was made of
emissions data and stack conditions (height, temperature, volume flow rate) for each source.
2.3 Representing road sources
Within 500m of a road, where strong concentration gradients exist, a highly detailed treatment of road
sources is required. In KCLurban we modelled road emissions as a series of road links 10 m long and
based on geographically accurate Ordnance Survey road map data. To reflect the important separation
of two or more lanes of traffic, on large roads/motorways, each side of the road was modelled
separately. Six road categories (and associated kernels) were used in London, including open roads
(motorway), typical roads (average urban roads surrounded by low rise buildings) and 4 types of street
canyon (classified by their orientation: north-south, east-west, southwest-northeast and southeastnorthwest). The “typical roads” had a road width of 20m and a building height of 10m and “street
canyons” had a width of 30m and a building height of 25m. The “street canyon” width and height details
were manually sampled from the 3D building model in London
http://www.casa.ucl.ac.uk/news/newsStory.asp?ID=80. There are approximately 2.25 million 10m road
sources in the KCLurban model.
2.4 Representing railway sources
Railway sources were treated in much the same way as for roads, i.e. by using the rail network emissions
broken into 10m sections. Whilst trains in London are predominantly electric, the line running west
from Paddington station and the line running north from St Pancras station includes many diesel trains
and represents a significant source of NOX and PM emissions. For diesel trains, the emissions release
height was taken to be 5m and the NO2/NOX exhaust ratio was assumed to be similar to large HGVs at
13%.
2.5 Representing gas combustion sources
Gas combustion is a very important source of NOX in London3, and in the KCLurban model we have
included a detailed representation of the height of release and spatial distribution of gas sources as well
as the temporal change in emissions throughout the year. Through analysis of the 3D model of buildings
in London (http://www.casa.ucl.ac.uk/news/newsStory.asp?ID=80) the height of release from gas
sources was varied from 1m (domestic housing), through 30m for small commercial premises to 75m
for large commercial office buildings. Gas heating sources were represented spatially by points located
at 50m intervals throughout the minor road network and set back from the road by 20m, assuming that
the majority of buildings follow the road network. Model kernels representing the varied release
conditions from these sources were then applied to domestic and commercial gas emissions
summarised in 1 x 1km grids across London. Finally, a detailed treatment of the emissions variation by
hour of the day and month of the year was included in the model and was based upon UK gas use
statistics (http://www.decc.gov.uk/en/content/cms/statistics/energy_stats/source/gas/gas.aspx).
6
2.6 Representing Heathrow airport
At Heathrow airport, emissions from aircraft during approach, landing, taxi out, taxi in, hold, take off,
initial climb and climb out were represented using the same method as for roads, i.e. as individual
sources 10m apart. Auxiliary Power Unit (APU) emissions and engine testing on the airport were
represented as stationary point sources. Other sources such as heating plant, public and staff car parks,
car rental, taxis queues and fire training ground emissions were represented horizontally as volume
sources of 1 x 1 km, 50m high, and for airside vehicles 2m high.
Takeoff is the aircraft mode that provides one of the largest contributions to ground-level NOX and PM
concentrations. During takeoff each accelerating aircraft engine was represented by horizontal
stationary jet sources at 10 m intervals along the runway, from the start of the ground-roll to where the
aircraft leaves the ground. The effect of accelerating aircraft was included to reflect the maximum
emission close to the start of the ground-roll, thereby reproducing the steep emission gradient along the
runway.
Of importance in estimating ground level concentrations from Jet plumes is their rapid change in
temperature and velocity over short distances, as well as the effect of aircraft speed. By comparing
modelled and measured data close to Heathrow airport we showed that buoyant jet plumes could be
accurately modelled by varying the jet release temperature assumptions. The hourly variation of aircraft
emissions was reproduced using aircraft movements made available by the UK Civil Aviation Authority.
Jet velocities were varied for the different aircraft operational settings of take off, approach and taxiing
and were assumed to be 85 %, 30 % and 7 % of full thrust, respectively. Finally, account was taken of the
rapidly reducing effect of aircraft emissions to ground level concentrations at different aircraft heights.
The aircraft NO2/NOX emissions ratio was assumed to be 11%.
2.7 Predicting annual mean NO2 concentrations
The method for converting NOX to NO2 used well established relationships of Carslaw et al7and was
based upon an analysis of measurements at both background and roadside sites. The conversion of NOX
to NO2 also included the influence of NO2 emitted directly from the vehicle exhaust.
2.8 References
1. CERC, 2004, ADMS 4 and Roads user guides. Available at www.cerc.co.uk.
2. CERC. 2006: ADMS-Roads An Air Quality Management System User Guide. Cambridge Environmental Research
Consultants Ltd, Cambridge. (http://www.cerc.co.uk/environmentalsoftware/assets/data/doc_userguides/CERC_ADMS-Roads%20Extra3.1_User_Guide.pdf). Accessed
03/01/2012
3. Greater London Authority (GLA). 2010. The London Atmospheric Emissions Inventory 2008.
http://data.london.gov.uk/laei-2008 Accessed 31/05/2011.
7
4. Fuller, G., Carslaw, D.C., Lodge, H.W., 2002. An empirical approach for the prediction of daily mean PM10
concentrations. Atmospheric Environment 36, 1431-1441.
5. Fuller, G.W. and Green, D., 2004. The Impact of local fugitive PM 10 from building works and road works on the
assessment of the European Union Limit Value. Atmospheric Environment 38, 4993-5002.
6. Fuller GW, Green D. 2006. Evidence for increasing concentrations of primary PM 10 in London. Atmospheric
Environment 40, 32, 6134-6145.
7. Carslaw D.C., Beevers S.D., Fuller G.W., 2001 An empirical approach for the prediction of annual mean nitrogen
dioxide concentrations in London, Atmospheric Environment, 35(8): 1505-1515.
8