2.1. Vehicle Emission Remote Sensing

The development of and operating principles behind vehicle emission remote sensing has been described in considerable detail in other publications,^19,20 but is summarized here. A remote sensing device (RSD) consists of a UV/IR source, multiple detectors, optical speed-acceleration bars, and a number plate camera. A RSD is deployed such that vehicles drive past the set-up unimpeded, with the concentrations of gases in their exhaust plumes and their speed and acceleration being measured remotely via open path spectroscopy. Spectrometry is achieved using a collinear beam of IR and UV light which, after being absorbed by exhaust plumes, is separated into its two components within the detector. Nondispersive infrared detectors measure CO, CO₂, hydrocarbons (HCs), and a background reference. The UV component passes through a quartz fiber bundle and is used to measure NH₃, NO, and NO₂.

One hundred measurements are taken in half a second for each vehicle plume exhaust when the rear of the vehicle is detected. From these measurements, the ratio of a pollutant to CO₂ is calculated, from which fuel-specific (g kg^–1) emission factors can be calculated. The further transformation from fuel-specific to distance-specific (g km^–1) emission factors is described later in the text.

Vehicle number plates are recorded alongside emission and speed measurements and are used to obtain vehicle technical data, such as engine size, fuel type, Euro standard, and vehicle manufacturer. In this study, the data were obtained from CDL Vehicle Information Services Ltd., a commercial supplier. CDL retrieved the data from the Driver and Vehicle Licensing Agency and the Society of Motor Manufacturers and Traders Motor Vehicle Registration Information System. Data relating to the total mileage of each vehicle at its last annual technical inspection test was also obtained through CDL for vehicles greater than three years old.

Vehicle emission measurements were conducted between 2017 and 2020 at 37 sites across 14 regions in the United Kingdom using two remote sensing instruments—the majority with the Opus AccuScan RSD 5000,²¹ supplemented with the data from the University of Denver Fuel Efficiency Automobile Test (FEAT) instrument.²² A total of 304,039 measurements were collected of Euro 2–6 vehicles in three key classes of LDVs: diesel light commercial vehicles (LCVs) and diesel and gasoline passenger cars (PCs). A statistical summary of the data set is provided in Table 1.