The hydroxyl radical (OH) represents the most important oxidant of the troposphere and the sink of most trace compounds in the atmosphere. If the sources of the OH radicals are nowadays relatively well constrained, the ‘sink’ term, also known as the total atmospheric reactivity with the OH radicals (or simply OH reactivity) (s-1), is still difficult to quantify. Due to the complexity of the used instruments, only few OH reactivity measurements are currently available. A new method proposed by Sinha et al. (2008) represents a promising alternative to the quantification of the total ambient OH reactivity. This method, also called the Comparative Reactivity Method (CRM), is based on the rapid measurement of a molecule normally not present in the ambient air (here, pyrrole, C4H5N) and which reacts at a known rate with OH radicals artificially generated into a glass reaction cell. Comparing the measured signals of pyrrole with and without ambient air within a constant OH field allows the quantification of the total OH reactivity of the ambient air. The main objective of this PhD work was the development and the improvement of this method for OH reactivity measurements in urban areas, which are abundant in nitrogen oxide (NO), a compound known to produce artifacts in the reactivity measurement. In the same time, it was necessary to propose a methodology to integrate all corrections (inherent to the method) to the raw reactivity data. Once the method adjusted, the second aim of this work was the characterization of the ambient OH reactivity in Paris, one of the few Megacities in Europe and where no other reactivity measurements were previously reported. The study conducted during the 2010 winter campaign of the MEGAPOLI European project allowed to quantify the OH reactivity level in Paris during two different air mass regimes, marine/continental. It was therefore possible to distinguish between a purely local OH reactivity level (about 20s-1) of Paris and Ile de France (impacted by the ‘traffic’ character of this European city and comparable to levels registered in New York and Tokyo) and a high (more than 100 s-1) , imported level characterized by a long distance transport. The study also revealed that during the continental import, high percentages (up to 75%) were registered of the missing OH reactivity, defined as the difference between the measured reactivity and a theoretical value obtained from the reactive compounds measured during the same campaign. The nature of the non-measured species contributing to this missing reactivity was determined as oxidized, issued of the air masses “processed” during the long distance transport. This work brings unique information on the CRM method and contributes to the characterization of the OH reactivity levels in the region of Paris.
