Pollution de l'air Dioxyde d'azote

1. What is Nitrogen Dioxide (NO₂)?

The source document for this Digest states:

As for PM and O₃, the evidence on NO₂ and health comes from different sources of information, including observational epidemiology, controlled human exposures to pollutants and animal toxicology. The observational data are derived from studies outdoors where NO₂ is one component of the complex mixture of different pollutants found in ambient air and from studies of NO₂ exposure indoors where its sources include unvented combustion appliances. Interpretation of evidence on NO₂ exposures outdoors is complicated by the fact that in most urban locations, the nitrogen oxides that yield NO₂ are emitted primarily by motor vehicles, making it a strong indicator of vehicle emissions (including other unmeasured pollutants emitted by these sources). NO₂ (and other nitrogen oxides) is also a precursor for a number of harmful secondary air pollutants, including nitric acid, the nitrate part of secondary inorganic aerosols and photo oxidants (including ozone). The situation is also complicated by the fact that photochemical reactions take some time (depending on the composition of the atmosphere and meteorological parameters) and air can travel some distance before secondary pollutants are generated. These relationships are shown schematically in Figure 1.

Figure 1: Simplified relationship of nitrogen oxides emissions with formation of NO₂ and other harmful reaction products including O₃ and PM

Health risks from nitrogen oxides may potentially result from NO₂ itself or its reaction products including O₃ and secondary particles. Epidemiological studies of NO₂ exposures from outdoor air are limited in being able to separate these effects. Additionally, NO₂ concentrations closely follow vehicle emissions in many situations so that NO₂ levels are generally a reasonable marker of exposure to traffic related emissions.

Given these complex relationships, findings of multivariate models that include NO₂ and other pollutants need cautious interpretation. While multi-pollutant models have been routinely applied to various forms of observational data, they may mis-specify underlying relationships. Even models that include only NO₂ and PM, NO₂ and O₃, or NO₂, PM and O₃ do not reflect the interrelationships among these pollutants. Statistical models considering interactions must be based on a strong a priori hypothesis about the nature of these interactions to allow their interpretation. With these constraints in mind, the working group recommended against using regression coefficients for NO₂ from regression models for the purpose of quantitative risk assessment.

Evidence of the health effects of NO₂ by itself thus comes largely from toxicological studies and from observational studies on NO₂ exposure indoors. The studies of outdoor NO₂ may be most useful under the following circumstances:

• Evidence for NO₂ effects assessed at fixed levels of exposure to other pollutants
• Evidence for modification of the effect of PM by NO₂, possibly indicating a potential consequence of HNO₃ vapour and/or PM nitrate.

Source & ©: WHO Regional Office for Europe  "Health Aspects of Air Pollution" (2003), Chapter 7 Nitrogen dioxide, Section 7.1 Introduction

The same information on