Real-World Emissions

At the current time there is considerable interest in air pollution; hardly a day passes without more evidence being published highlighting the serious effects air pollution has on health.

The UK has a long and interesting history of air pollution dating back centuries. In the 1950s and 1960s the focus was very much on controlling domestic coal combustion. In urban areas today however, the focus is on the impact that road vehicles have on air pollution. For many pollutants, the emissions from road vehicles have either been eliminated (such as lead) or reduced to very low levels (such as carbon monoxide and sulphur dioxide) that they are no longer a concern.

However, it has been much more challenging to reduce nitrogen oxides (NO_x). While the emission of NO_x from road vehicles has reduced over the past decade, there remain many locations, mostly close to roads, that exceed EU Directive limits for nitrogen dioxide (NO₂). The principal challenge is to meet the annual mean Limit Value of 40 µm-₃. This is also a problem shared by many other European countries – a consequence of similar vehicle fleets across Europe.

Currently in the UK there are around 60 local authorities considering Clean Air Zones (CAZ) or Low Emission Zones (LEZ) to accelerate the delivery of cleaner air. Nearly all of these zones use vehicle emission Euro standards as a way of restricting certain vehicles into particular areas of cities. It is well documented that the emissions performance of vehicles on the road can be very different from their certified Euro standard emissions levels. These discrepancies threaten to undermine well-intentioned pollution reduction policies, resulting in interventions that fail to deliver the benefits originally anticipated.

Vehicle emissions in the real world

Over the last few years, extensive vehicle emission measurements have been made using a technique based on remote sensing. Ricardo’s database of vehicle emissions of over 300,000 measurements; for example, accurately records the real-world driving emissions from vehicles, at the roadside, under actual driving conditions. It is interesting that in the field of vehicle emissions, the term ‘real-world emissions’ is often used. This is because almost all measurements of vehicle emissions are laboratory-based and made on rolling roads, e.g. as part of the Type Approval process. It is, however, challenging to measure emissions from vehicles under actual conditions of use with ‘real’ drivers driving their own vehicles.

One of the advantages of remote sensing measurements is that they directly provide fleet-weighted average real-world emissions. An important strength of this approach is that it provides insight into the whole vehicle fleet. In principle, this allows for the quantification of emissions from vehicles directly affected by a CAZ and those that are not. Both pieces of this jigsaw are needed for a robust understanding of what any policy will bring regarding vehicle emissions reductions

Figure 1. Vehicle emissions remote sensing requires no contact with the vehicle being measured

Real-world vehicle emissions databases provide estimates of the g km-₁ of nitrogen oxides (NO_x) by vehicle type, Euro standard and vehicle model. With about 100,000 measurements of diesel cars alone, it is possible to provide a robust indication of how emissions have changed through the different Euro standard. The data are also comprehensive enough to predict emissions over a wide range of driving conditions and driving cycles. Figure 1 shows how real-world vehicle measurements are gathered directly from the roadside.

Figure 2 shows how emissions of NO_x have changed in going from Euro 2 to Euro 6 vehicles – spanning a period from the mid 1990s to the present day. The results show that emissions of NO_x did not change by much until Euro 6 vehicles were introduced around 2013. The measurements also show there is a large difference between the two main Euro 6 technologies: Lean NO_x Traps (LNT) and Selective Catalytic Reduction (SCR). LNT tends to be used on smaller vehicles and SCR on larger vehicles. It is clear from the measurements that SCR technology significantly reduces emissions of NO_x compared with LNT.

With so many measurements it is possible to unpick the emissions data in a myriad of ways. We have, for example, considered effects such as emissions deterioration and ambient temperature – issues for which there is little data available. However, one interesting issue to address is how much variation is there in emissions within one Euro standard – and what might the implications be for mitigation of emissions through the use of Euro standard restrictions?

Figure 3 shows the emissions from Euro 5 diesel cars split by manufacturer. This plot shows there is a large difference between manufacturers of about a factor of three. Indeed, there is an even larger distribution of emissions if the data are further split by vehicle model or engine size. Considering Euro 5 vehicles as a homogenous group does not reflect the important variation in emissions performance within the Euro 5 standard.

What about the situation for Euro 6 diesel passenger cars?

The emissions from these vehicles are shown in Figure 4, plotted on the same scale as Figure 3. What is apparent is that overall, the emissions of NO_x from Euro 6 vehicles tend to be much lower than Euro 5. However, there are several important aspects of Figure 4 that should be noted. First, there is a clear separation between the performance of SCR and LNT-equipped vehicles with SCR tending to have much lower emissions. Second, the difference between ‘best’ to ‘worst’ is much greater than for Euro 5 cars. Indeed, the differential between best and worst is even greater when individual models of vehicle are considered.

A comparison can also be made to see how emissions of NO_x changed in going from Euro 5 to Euro 6 for each manufacturer, as shown in Figure 5.

In this Figure it can be seen that nearly all Euro 6 vehicles show a decrease in NO_x emissions relative to Euro 5, as indicated by points being below the dashed line. The size of the circles represents the number of vehicles a manufacturer has in the fleet. The larger circles therefore highlight those manufacturers that exert most influence on the overall NO_x emissions. Those circles closer to the 1:1 (dashed line) represent vehicle manufacturers that have not reduced NO_x as effectively as what is technically possible. Indeed, large circles close to the 1:1 line represent manufacturers that have a disproportionate impact on overall fleet reductions of NO_x.

The differential performance in NO_x emissions shown in the above plots highlights that Euro class alone does not guarantee a particular emissions performance. However, the fleet average Euro 6 emissions of NO_x are lower than Euro 5. In this respect, measures that aim to restrict pre-Euro 6 vehicles will result in lower emissions of NO_x overall. From an emissions mitigation perspective, this means that some older, but low emission vehicles, could be barred from entering LEZs or CAZs. This reality has consequences for the optimum design of emissions reduction policies, arguably raising the question as to whether schemes based on Euro standards are too blunt an approach to effectively tackle air pollution.

On the 8th April 2019, a new Ultra Low Emission Zone (ULEZ) was introduced in central London in the same area as the Congestion Charge. The initial stage of the London ULEZ will cover the same area as that of the existing Congestion Charge. Here, understanding the whole fleet is important. This zone has a very different fleet composition compared with that for the rest of London, including a higher proportion of Transport for London (TfL) buses and taxis than the wider London or UK fleet. These vehicles are important contributors to NO_x emissions, and NO₂ in particular, in the zone. Data collected, in partnership with the International Council on Clean Transportation (Dallmann et al. 2018), highlights that good progress is being made in reducing NO_x emissions from buses with big gains accrued from the move to Euro 6. However, the situation for taxis is far more challenging. Taxis, as high NO_x emitters, will continue to make an important contribution to central London emissions for some time to come. Ricardo’s real-world emissions measurements suggest that Euro 5 black cabs emit around 2.2 g km-₁ (i.e. at the high end of the distributions shown in Figure 3).

There are also other factors at play that are emerging from the latest evidence. For instance, recent analysis of ambient air quality concentration data from across Europe, coupled with insight delivered from our real-world emissions data, is highlighting the important role of primary NO₂ emissions (i.e. the NO₂ directly emitted by vehicles) in this mix. Even though it has been known that primary NO₂ emissions can make an important contribution to urban NO₂ concentrations, the emissions have been somewhat ignored. In some respects, this is because vehicle emissions legislation itself only considers total NO_x and is not concerned about how much of the NO_x is NO₂.

Conversely, ambient limits set for health protection, e.g. the annual mean Limit Value of 40 µ m-₃, considers NO₂ and not total NO_x. From the perspective of ambient NO₂ concentrations and impacts on health, it is clear a consideration of emissions should consider both total NO_x and the amount emitted as NO₂.

The good news is that primary NO₂ emissions have reduced significantly since 2010 (Carslaw et al. 2019; Grange et al. 2017). This reduction is driven by a several factors but it includes improved control of NO₂ emissions from new vehicles and the fact that diesel cars emit less NO₂ as their mileage increases (Carslaw et al. 2019). However, much of the work that has been done in developing CAZ/ LEZs has not accounted for these changes. This evidence has the potential to be an important insight as it is possible, maybe likely, that emissions and concentrations of NO₂ will decrease more rapidly than is currently predicted.

Understanding real-world emissions and other related factors is critical to the development of well-informed air quality interventions such as CAZs. In this respect, carefully analysing ambient air-quality concentration is of great importance to ensure that changes in emissions have the anticipated effects on the actual concentrations of the pollutants that we breathe. After all, it is the concentration of pollutants, not the emissions themselves (although clearly linked), that must be addressed if the health of those exposed is to improve. What matters, and what we really need, is a low concentration zone (or a ‘low exposure zone’ – also known as a LEZ).

Ricardo’s ongoing programme of measuring real-world driving emissions continues to build what is now a unique database of almost 350,000 individual vehicle emissions measurements – the largest database of real-world emissions information in the UK. Ricardo is currently involved in work to understand the impact of the introduction of the ULEZ in London, as well as assessing the impact of a Bus Emissions Abatement Retrofit Programme in Scotland, supporting local authorities in developing and understanding CAZs.

References
Carslaw, David C. et al. 2019. “The Diminishing Importance of Nitrogen Dioxide Emissions from Road Vehicle Exhaust.” Atmospheric Environment: X 1(2): 100002. https://www.sciencedirect.com/science/article/pii/S2590162118300029.
Dallmann, Tim, Yoann Bernard, Uwe Tietge, and Rachel Muncrief. 2018. Remote Sensing of Motor Vehicle Emissions in London. https://www.theicct.org/publications/ true-london-dec2018.
Grange, S.K., A.C. Lewis, S.J. Moller, and D.C. Carslaw. 2017. “Lower Vehicular Primary Emissions of NO2 in Europe than Assumed in Policy Projections.” Nature Geoscience 10(12).