Indoor cooking and cleaning as a source of outdoor air pollution in urban environments.

Indoor sources of air pollution, such as from cooking and cleaning, play a key role in indoor gas-phase chemistry. The focus of the impact of these activities on air quality tends to be indoors, with less attention given to the impact on air quality outside buildings. This study uses the INdoor CHEmical Model in Python (INCHEM-Py) and the Advanced Dispersion Modelling System (ADMS) to quantify the impact cooking and cleaning have on indoor and outdoor air quality for an idealised street of houses. INCHEM-Py has been developed to determine the concentrations of 106 indoor volatile organic compounds at the point they leave a building (defined as near-field concentrations). For a simulated 140 m long street with 10 equi-distant houses undertaking cooking and cleaning activities, the maximum downwind concentration of acetaldehyde increases from a background value of 0.1 ppb to 0.9 ppb post-cooking, whilst the maximum downwind chloroform concentrations increase from 1.2 to 6.2 ppt after cleaning. Although emissions to outdoors are higher when cooking and cleaning happen indoors, the contribution of these activities to total UK emissions of volatile organic compounds is low (less than 1%), and comprise about a quarter of those emitted from traffic across the UK. It is important to quantify these emissions, particularly as continued vehicle technology improvements lead to lower direct emissions outdoors, making indoor emissions relatively more important. Understanding how indoor pollution can affect outdoor environments, will allow better mitigation measures to be designed in the future that can take into account all sources of pollution that contribute to human exposure.

The impact on passenger car emissions associated with the promotion and demise of diesel fuel.

The promotion and growth in the use of diesel fuel in passenger cars in the UK and Europe over the past two decades led to considerable adverse air quality impacts in urban areas and more widely. In this work, we construct a multi-decade analysis of passenger car emissions in the UK based on real driving emissions data. An important part of the study is the use of extensive vehicle emission remote sensing data covering multiple measurement locations, time periods, environmental conditions and consisting of over 600,000 measurements. These data are used to consider two scenarios: first, that diesel fuel use was not promoted in the early 2000s for climate mitigation reasons, and second, that there was not a dramatic decline in diesel fuel use following the Dieselgate scandal. The strong growth of diesel fuel use coincided with a time when diesel NOx emissions were high and, conversely, the strong decrease of diesel fuel use coincided with a time when diesel vehicle after-treatment systems for NOx control were effective. We estimate that the growth in diesel car use in the UK results in excess NOx emissions of 721 kt over a three decade period; equivalent to over 7 times total annual passenger car NOx emissions and greater than total UK NOx emissions of 681.8 kt in 2021 and with an associated damage cost of £5.875 billion. However, the sharp move away from diesel fuel post-Dieselgate only reduced NOx emissions by 41 kt owing to the effectiveness of modern diesel aftertreatment systems.

Emission ratio determination from road vehicles using a range of remote emission sensing techniques.

The development of remote emission sensing techniques such as plume chasing and point sampling has progressed significantly and is providing new insight into vehicle emissions behaviour. However, the analysis of remote emission sensing data can be highly challenging and there is currently no standardised method available. In this study we present a single data processing approach to quantify vehicle exhaust emissions measured using a range of remote emission sensing techniques. The method uses rolling regression calculated over short time intervals to derive the characteristics of diluting plumes. We apply the method to high time-resolution plume chasing and point sampling data to quantify gaseous exhaust emission ratios from individual vehicles. Data from a series of vehicle emission characterisation experiments conducted under controlled conditions is used to demonstrate the potential of this approach. First, the method is validated through comparison with on-board emission measurements. Second, the ability of this approach to detect changes in NOx / CO2 ratios associated with aftertreatment system tampering and different engine operating conditions is shown. Third, the flexibility of the approach is demonstrated by varying the pollutants used as regression variables and quantifying the NO2 / NOx ratios for different vehicle types. A higher proportion of total NOx is emitted as NO2 when the selective catalytic reduction system of the measured heavy duty truck is tampered. In addition, the applicability of this approach to urban environments is illustrated using mobile measurements conducted in Milan, Italy in 2021. Emissions from local combustion sources are distinguished from a complex urban background and the spatiotemporal variability in emissions is shown. The mean NOx / CO2 ratio of 1.61 ppb/ppm is considered representative of the local vehicle fleet. It is envisaged that this approach can be used to quantify emissions from a range of mobile and stationary fuel combustion sources, including non-road vehicles, ships, trains, boilers and incinerators.

Can accurate distance-specific emissions of nitrogen oxide emissions from cars be determined using remote sensing without measuring exhaust flowrate?

Portable Emission Measurement Systems (PEMS) are commonly used to measure absolute (mass per unit distance) emissions of a range of pollutants from road vehicles under real driving conditions. Because measuring large numbers of vehicles with PEMS is impractical, this paper investigates how vehicle emission remote sensing device (RSD) can supplement the use of PEMS. We simulate whether remote sensing measurements can accurately predict a vehicle's real-world distance-specific nitrogen oxides (NOX) emissions using RSD without measuring its exhaust flow rate. The approach uses readily available type-approval carbon dioxide (CO2) emission data together with average real-world divergences from studies based on user-reported fuel economy data. We find that at least 30 RS measurements from a given vehicle's journey are needed to reach a mean absolute error of 30% compared to a large reference data set of individual PEMS measurements. With that condition met, it is concluded that estimates agree well with actual NOX emissions from cars and the applied method does not introduce a systematic bias. It is also found that the accuracy of estimates for distance-specific NOX emissions does not significantly improve when more than 300 remote-sensing samples are available, with a mean absolute error converging to 23%. We conclude that this method could be used to screen large car fleets and identify vehicles or group of vehicles that are likely grossly exceeding air pollution standards.

Meteorological Normalisation Using Boosted Regression Trees to Estimate the Impact of COVID-19 Restrictions on Air Quality Levels.

The global COVID-19 pandemic that began in late December 2019 led to unprecedented lockdowns worldwide, providing a unique opportunity to investigate in detail the impacts of restricted anthropogenic emissions on air quality. A wide range of strategies and approaches exist to achieve this. In this paper, we use the "deweather" R package, based on Boosted Regression Tree (BRT) models, first to remove the influences of meteorology and emission trend patterns from NO, NO2, PM10 and O3 data series, and then to calculate the relative changes in air pollutant levels in 2020 with respect to the previous seven years (2013-2019). Data from a northern Spanish region, Cantabria, with all types of monitoring stations (traffic, urban background, industrial and rural) were used, dividing the calendar year into eight periods according to the intensity of government restrictions. The results showed mean reductions in the lockdown period above -50% for NOx, around -10% for PM10 and below -5% for O3. Small differences were found between the relative changes obtained from normalised data with respect to those from observations. These results highlight the importance of developing an integrated policy to reduce anthropogenic emissions and the need to move towards sustainable mobility to ensure safer air quality levels, as pre-existing concentrations in some cases exceed the safe threshold.

Verification of a National Emission Inventory and Influence of On-road Vehicle Manufacturer-Level Emissions.

Road vehicles make important contributions to a wide range of pollutant emissions from the street level to global scales. The quantification of emissions from road vehicles is, however, highly challenging given the number of individual sources involved and the myriad factors that influence emissions such as fuel type, emission standard, and driving behavior. In this work, we use highly detailed and comprehensive vehicle emission remote sensing measurements made under real driving conditions to develop new bottom-up inventories that can be compared to official national inventory totals. We find that the total UK passenger car and light-duty van emissions of nitrogen oxides (NOx) are underestimated by 24-32%, and up to 47% in urban areas, compared with the UK national inventory, despite agreement within 1.5% for total fuel used. Emissions of NOx at a country level are also shown to vary considerably depending on the mix of vehicle manufacturers in the fleet. Adopting the on-road mix of vehicle manufacturers for six European countries results in up to a 13.4% range in total emissions of NOx. Accounting for the manufacturer-specific fleets at a country level could have a significant impact on emission estimates of NOx and other pollutants across the European countries, which are not currently reflected in emission inventories.

Underestimated Ammonia Emissions from Road Vehicles.

In this study, we use comprehensive vehicle emission remote sensing measurements of over 230,000 passenger cars to estimate total UK ammonia (NH3) emissions. Estimates are made using "top-down" and "bottom-up" methods that demonstrate good agreement to within 1.1% for total fuel consumed or CO2 emitted. A central component of this study is the comprehensive nature of the bottom-up emission estimates that combine highly detailed remote sensing emission data with over 4000 km of 1 Hz real driving data. Total annual UK NH3 emissions from gasoline passenger cars are estimated to be 7.8 ± 0.3 kt from the bottom-up estimate compared with 3.0 ± 1.7 kt reported by the UK national inventory. An important conclusion from the analysis is that both methodologies confirm that gasoline passenger car NH3 emissions are underestimated by a factor of about 2.6 compared with the 2018 UK National Atmospheric Emissions Inventory. Furthermore, we find that inventory estimates of urban emissions of NH3 for passenger cars are underestimated by a factor of 17.

An increasing role for solvent emissions and implications for future measurements of volatile organic compounds.

Volatile organic compounds (VOCs) are a broad class of air pollutants which act as precursors to tropospheric ozone and secondary organic aerosols. Total UK emissions of anthropogenic VOCs peaked in 1990 at 2,840 kt yr-1 and then declined to approximately 810 kt yr-1 in 2017 with large reductions in road transport and fugitive fuel emissions. The atmospheric concentrations of many non-methane hydrocarbons (NMHC) in the UK have been observed to fall over this period in broadly similar proportions. The relative contribution to emissions from solvents and industrial processes is estimated to have increased from approximately 35% in 1990 to approximately 63% in 2017. In 1992, UK national monitoring quantified 19 of the 20 most abundant individual anthropogenic VOCs emitted (all were NMHCs), but by 2017 monitoring captured only 13 of the top 20 emitted VOCs. Ethanol is now estimated to be the most important VOC emitted by mass (in 2017 approx. 136 kt yr-1 and approx. 16.8% of total emissions) followed by n-butane (52.4 kt yr-1) and methanol (33.2 kt yr-1). Alcohols have grown in significance representing approximately 10% of emissions in 1990 rising to approximately 30% in 2017. The increased role of solvent emissions should now be reflected in European monitoring strategies to verify total VOC emission reduction obligations in the National Emissions Ceiling Directive. Adding ethanol, methanol, formaldehyde, acetone, 2-butanone and 2-propanol to the existing NMHC measurements would provide full coverage of the 20 most significant VOCs emitted on an annual mass basis. This article is part of a discussion meeting issue 'Air quality, past present and future'.

Distance-based emission factors from vehicle emission remote sensing measurements.

Vehicle emission remote sensing has the potential to provide detailed emissions information at a highly disaggregated level owing to the ability to measure thousands of vehicles in a single day. Fundamentally, vehicle emission remote sensing provides a direct measure of the molar volume ratio of a pollutant to carbon dioxide, from which fuel-based emissions factors can readily be calculated. However, vehicle emissions are more commonly expressed in emission per unit distance travelled e.g. grams per km or mile. To express vehicle emission remote sensing data in this way requires an estimate of the fuel consumption at the time of the emission measurement. In this paper, an approach is developed based on vehicle specific power that uses commonly measured or easily obtainable vehicle information such as vehicle speed, acceleration and mass. We test the approach against 55 independent comprehensive PEMS measurements for Euro 5 and 6 gasoline and diesel vehicles over a wide range of driving conditions and find good agreement between the method and PEMS data. The method is applied to individual vehicle model types to quantify distance-based emission factors. The method will be appropriate for application to larger vehicle emission remote sensing databases, thus extending real-world distance-based vehicle emissions information.

Post-Dieselgate: Evidence of NOx Emission Reductions Using On-Road Remote Sensing.

The Dieselgate scandal which broke in September 2015 demonstrated that vehicle manufacturers, such as the Volkswagen Group (VWG), engaged in software-based manipulation which led to vehicles passing laboratory-based emission testing limits but were far more polluting while being driven on roads. Using 23 000 on-road remote sensing measurements of light-duty Euro 5 diesel vehicles in the United Kingdom between 2012 and 2018, VWG vehicles with the "Dieselgate-affected" EA189 engine demonstrated anomalous NOx emission behavior between the pre- and post-Dieselgate periods which was not observed in other vehicle makes or models. These anomalous changes can be explained by voluntary VWG hardware and software fixes which have led to improved NOx emission control. The VGW 1.6 L vehicles, with a simple hardware fix and a software upgrade, resulted in a 36% reduction in NOx, whereas the 2.0 L vehicles that required a software-only fix showed a 30% reduction in NOx once controlled for ambient temperature effects. These results show that even minor changes or upgrades can considerably reduce NOx emissions, which has implications for future emission control activities and local air quality.

Strong Temperature Dependence for Light-Duty Diesel Vehicle NO x Emissions.

Diesel-powered road vehicles are important sources for nitrogen oxide (NO x) emissions, and the European passenger fleet is highly dieselised, which has resulted in many European roadside environments being noncompliant with legal air quality standards for nitrogen dioxide (NO2). On the basis of vehicle emission remote sensing data for 300000 light-duty vehicles across the United Kingdom, light-duty diesel NO x emissions were found to be highly dependent on ambient temperature with low temperatures resulting in higher NO x emissions, i.e., a "low temperature NO x emission penalty" was identified. This feature was not observed for gasoline-powered vehicles. Older Euro 3 to 5 diesel vehicles emitted NO x similarly, but vehicles compliant with the latest Euro 6 emission standard emitted less NO x than older vehicles and demonstrated less of an ambient temperature dependence. This ambient temperature dependence is overlooked in current emission inventories but is of importance from an air quality perspective. Owing to Europe's climate, a predicted average of 38% more NO x emissions have burdened Europe when compared to temperatures encountered in laboratory test cycles. However, owing to the progressive elimination of vehicles demonstrating the most severe low temperature NO x penalty, light-duty diesel NO x emissions are likely to decrease more rapidly throughout Europe than currently thought.

Using meteorological normalisation to detect interventions in air quality time series.

Interventions used to improve air quality are often difficult to detect in air quality time series due to the complex nature of the atmosphere. Meteorological normalisation is a technique which controls for meteorology/weather over time in an air quality time series so intervention exploration (and trend analysis) can be assessed in a robust way. A meteorological normalisation technique, based on the random forest machine learning algorithm was applied to routinely collected observations from two locations where known interventions were imposed on transportation activities which were expected to change ambient pollutant concentrations. The application of progressively stringent limits on the content of sulfur in marine fuels was very clearly represented in ambient sulfur dioxide (SO2) monitoring data in Dover, a port city in the South East of England. When the technique was applied to the oxides of nitrogen (NOx and NO2) time series at London Marylebone Road (a Central London monitoring site located in a complex urban environment), the normalised time series highlighted clear changes in NO2 and NOx which were linked to changes in primary (directly emitted) NO2 emissions at the location. The clear features in the time series were illuminated by the meteorological normalisation procedure and were not observable in the raw concentration data alone. The lack of a need for specialised inputs, and the efficient handling of collinearity and interaction effects makes the technique flexible and suitable for a range of potential applications for air quality intervention exploration.

Interinstrument comparison of remote-sensing devices and a new method for calculating on-road nitrogen oxides emissions and validation of vehicle-specific power.

Emissions of nitrogen oxides (NOx) by vehicles in real driving environments are only partially understood. This has been brought to the attention of the world with recent revelations of the cheating of the type of approval tests exposed in the dieselgate scandal. Remote-sensing devices offer investigators an opportunity to directly measure in situ real driving emissions of tens of thousands of vehicles. Remote-sensing NO2 measurements are not as widely available as would be desirable. The aim of this study is to improve the ability of investigators to estimate the NO2 emissions and to improve the confidence of the total NOx results calculated from standard remote-sensing device (RSD) measurements. The accuracy of the RSD speed and acceleration module was also validated using state-of-the-art onboard global positioning system (GPS) tracking. Two RSDs used in roadside vehicle emissions surveys were tested side by side under off-carriageway conditions away from transient pollution sources to ascertain the consistency of their measurements. The speed correlation was consistent across the range of measurements at 95% confidence and the acceleration correlation was consistent at 95% confidence intervals for all but the most extreme acceleration cases. VSP was consistent at 95% confidence across all measurements except for those at VSP ≥ 15 kW t-1, which show a small underestimate. The controlled distribution gas nitric oxide measurements follow a normal distribution with 2σ equal to 18.9% of the mean, compared to 15% observed during factory calibration indicative of additional error introduced into the system. Systematic errors of +84 ppm were observed but within the tolerance of the control gas. Interinstrument correlation was performed, with the relationship between the FEAT and the RSD4600 being linear with a gradient of 0.93 and an R2 of 0.85, indicating good correlation. A new method to calculate NOx emissions using fractional NO2 combined with NO measurements made by the RSD4600 was constructed, validated, and shown to be more accurate than previous methods. IMPLICATIONS: Synchronized remote-sensing measurements of NO were taken using two different remote-sensing devices in an off-road study. It was found that the measurements taken by both instruments were well correlated. Fractional NO2 measurements from a prior study, measurable on only one device, were used to create new NOx emission factors for the device that could not be measured by the second device. These estimates were validated against direct measurement of total NOx emission factors and shown to be an improvement on previous methodologies. Validation of vehicle-specific power was performed with good correlation observed.

Spatially resolved flux measurements of NOx from London suggest significantly higher emissions than predicted by inventories.

To date, direct validation of city-wide emissions inventories for air pollutants has been difficult or impossible. However, recent technological innovations now allow direct measurement of pollutant fluxes from cities, for comparison with emissions inventories, which are themselves commonly used for prediction of current and future air quality and to help guide abatement strategies. Fluxes of NOx were measured using the eddy-covariance technique from an aircraft flying at low altitude over London. The highest fluxes were observed over central London, with lower fluxes measured in suburban areas. A footprint model was used to estimate the spatial area from which the measured emissions occurred. This allowed comparison of the flux measurements to the UK's National Atmospheric Emissions Inventory (NAEI) for NOx, with scaling factors used to account for the actual time of day, day of week and month of year of the measurement. The comparison suggests significant underestimation of NOx emissions in London by the NAEI, mainly due to its under-representation of real world road traffic emissions. A comparison was also carried out with an enhanced version of the inventory using real world driving emission factors and road measurement data taken from the London Atmospheric Emissions Inventory (LAEI). The measurement to inventory agreement was substantially improved using the enhanced version, showing the importance of fully accounting for road traffic, which is the dominant NOx emission source in London. In central London there was still an underestimation by the inventory of 30-40% compared with flux measurements, suggesting significant improvements are still required in the NOx emissions inventory.

Have vehicle emissions of primary NO2 peaked?

Reducing ambient concentrations of nitrogen dioxide (NO2) remains a key challenge across many European urban areas, particularly close to roads. This challenge mostly relates to the lack of reduction in emissions of oxides of nitrogen (NOx) from diesel road vehicles relative to the reductions expected through increasingly stringent vehicle emissions legislation. However, a key component of near-road concentrations of NO2 derives from directly emitted (primary) NO2 from diesel vehicles. It is well-established that the proportion of NO2 (i.e. the NO2/NOx ratio) in vehicle exhaust has increased over the past decade as a result of vehicle after-treatment technologies that oxidise carbon monoxide and hydrocarbons and generate NO2 to aid the emissions control of diesel particulate. In this work we bring together an analysis of ambient NOx and NO2 measurements with comprehensive vehicle emission remote sensing data obtained in London to better understand recent trends in the NO2/NOx ratio from road vehicles. We show that there is evidence that NO2 concentrations have decreased since around 2010 despite less evidence of a reduction in total NOx. The decrease is shown to be driven by relatively large reductions in the amount of NO2 directly emitted by vehicles; from around 25 vol% in 2010 to 15 vol% in 2014 in inner London, for example. The analysis of NOx and NO2 vehicle emission remote sensing data shows that these reductions have been mostly driven by reduced NO2/NOx emission ratios from heavy duty vehicles and buses rather than light duty vehicles. However, there is also evidence from the analysis of Euro 4 and 5 diesel passenger cars that as vehicles age the NO2/NOx ratio decreases. For example the NO2/NOx ratio decreased from 29.5 ± 2.0% in Euro 5 diesel cars up to one year old to 22.7 ± 2.5% for four-year old vehicles. At some roadside locations the reductions in primary NO2 have had a large effect on reducing both the annual mean and number of hourly exceedances of the European Limit Values of NO2.

Crucial Role for Outdoor Chemistry in Ultrafine Particle Formation in Modern Office Buildings.

In the developed world, we spend most of our time indoors, where we receive the majority of our exposure to air pollution. This paper reports model simulations of PM2.5 and ozone concentrations in identical landscape offices in three European cities: Athens, Helsinki, and Milan. We compare concentrations during an intense heatwave in August 2003 with a meteorologically more typical August in 2009. During the heatwave, average indoor ozone concentrations during office hours were 44, 19, and 41 ppb in Athens, Helsinki, and Milan respectively, enhanced by 7, 4, and 17 ppb respectively relative to 2009. Total predicted PM2.5 concentrations were 13.5, 3.6, and 17.2 µg m(-3) in Athens, Helsinki, and Milan respectively, enhanced by 0.5, 0.4, and 6.7 µg m(-3) respectively relative to 2009: the three cities were affected to differing extents by the heatwave. A significant portion of the indoor PM2.5 derived from gas-phase chemistry outdoors, producing 2.5, 0.8, and 4.8 µg m(-3) of the total concentrations in Athens, Helsinki, and Milan, respectively. Despite filtering office inlet supplies to remove outdoor particles, gas-phase precursors for particles can still enter offices, where conditions are ripe for new particles to form, particularly where biogenic emissions are important outdoors. This result has important implications for indoor air quality, particularly given the current trend for green walls on buildings, which will provide a potential source of biogenic emissions near to air inlet systems.

Measurement of NO(x) fluxes from a tall tower in Central London, UK and comparison with emissions inventories.

Direct measurements of NOx concentration and flux were made from a tall tower in central London, UK as part of the Clean Air for London (ClearfLo) project. Fast time resolution (10 Hz) NO and NO2 concentrations were measured and combined with fast vertical wind measurements to provide top-down flux estimates using the eddy covariance technique. Measured NOx fluxes were usually positive and ranged from close to zero at night to 2000-8000 ng m(-2) s(-1) during the day. Peak fluxes were usually observed in the morning, coincident with the maximum traffic flow. Measurements of the NOx flux have been scaled and compared to the UK National Atmospheric Emissions Inventory (NAEI) estimate of NOx emission for the measurement footprint. The measurements are on average 80% higher than the NAEI emission inventory for all of London. Observations made in westerly airflow (from parts of London where traffic is a smaller fraction of the NOx source) showed a better agreement on average with the inventory. The observations suggest that the emissions inventory is poorest at estimating NOx when traffic is the dominant source, in this case from an easterly direction from the BT Tower. Agreement between the measurements and the London Atmospheric Emissions Inventory (LAEI) are better, due to the more explicit treatment of traffic flow by this more detailed inventory. The flux observations support previous tailpipe observations of higher NOx emitted from the London vehicle diesel fleet than is represented in the NAEI or predicted for several EURO emission control technologies. Higher-than-anticipated vehicle NOx is likely responsible for the significant discrepancies that exist in London between observed NOx and long-term NOx projections.

Air pollution dispersion models for human exposure predictions in London.

The London household survey has shown that people travel and are exposed to air pollutants differently. This argues for human exposure to be based upon space-time-activity data and spatio-temporal air quality predictions. For the latter, we have demonstrated the role that dispersion models can play by using two complimentary models, KCLurban, which gives source apportionment information, and Community Multi-scale Air Quality Model (CMAQ)-urban, which predicts hourly air quality. The KCLurban model is in close agreement with observations of NO(X), NO(2) and particulate matter (PM)(10/2.5), having a small normalised mean bias (-6% to 4%) and a large Index of Agreement (0.71-0.88). The temporal trends of NO(X) from the CMAQ-urban model are also in reasonable agreement with observations. Spatially, NO(2) predictions show that within 10's of metres of major roads, concentrations can range from approximately 10-20 p.p.b. up to 70 p.p.b. and that for PM(10/2.5) central London roadside concentrations are approximately double the suburban background concentrations. Exposure to different PM sources is important and we predict that brake wear-related PM(10) concentrations are approximately eight times greater near major roads than at suburban background locations. Temporally, we have shown that average NO(X) concentrations close to roads can range by a factor of approximately six between the early morning minimum and morning rush hour maximum periods. These results present strong arguments for the hybrid exposure model under development at King's and, in future, for in-building models and a model for the London Underground.

One way coupling of CMAQ and a road source dispersion model for fine scale air pollution predictions.

In this paper we have coupled the CMAQ and ADMS air quality models to predict hourly concentrations of NO X , NO2 and O3 for London at a spatial scale of 20 m × 20 m. Model evaluation has demonstrated reasonable agreement with measurements from 80 monitoring sites in London. For NO2 the model evaluation statistics gave 73% of the hourly concentrations within a factor of two of observations, a mean bias of -4.7 ppb and normalised mean bias of -0.17, a RMSE value of 17.7 and an r value of 0.58. The equivalent results for O3 were 61% (FAC2), 2.8 ppb (MB), 0.15 (NMB), 12.1 (RMSE) and 0.64 (r). Analysis of the errors in the model predictions by hour of the week showed the need for improvements in predicting the magnitude of road transport related NO X emissions as well as the hourly emissions scaling in the model. These findings are consistent with recent evidence of UK road transport NO X emissions, reported elsewhere. The predictions of wind speed using the WRF model also influenced the model results and contributed to the daytime over prediction of NO X concentrations at the central London background site at Kensington and Chelsea. An investigation of the use of a simple NO-NO2-O3 chemistry scheme showed good performance close to road sources, and this is also consistent with previous studies. The coupling of the two models raises an issue of emissions double counting. Here, we have put forward a pragmatic solution to this problem with the result that a median double counting error of 0.42% exists across 39 roadside sites in London. Finally, whilst the model can be improved, the current results show promise and demonstrate that the use of a combination of regional scale and local scale models can provide a practical modelling tool for policy development at intergovernmental, national and local authority level, as well as for use in epidemiological studies.