A Detailed Study Calculating Hydroxyl Radical Concentration from Formaldehyde and VOC Measurements Made During the PEROXY Campaign, Cape Verde,2023

The hydroxyl radical, OH, is the major daytime oxidant in the troposphere, it controls the lifetime of methane and reacts with volatile organic compounds (VOCs), emitted from both anthropogenic and biogenic sources, often forming formaldehyde as a product. OH is technically difficult to measure, owing to its low atmospheric concentrations, but Wolfe et al., demonstrated that the variation in HCHO concentration can reflect the variation in OH, methane and VOC concentrations and using this relationship were able to estimate OH concentrations during ATom flights [1].

Taking measurements of HCHO made during the PEROXY campaign at the Cabo Verde observatory in 2023, this work presents, the effective yield of HCHO from all OH reactions (α) calculated by determining α from each VOC measured during the campaign and also using a detailed box model run with the Master Chemical Mechanism [2] and constrained to the observed VOCs to account for HCHO formed from OH reactions with the model generated intermediate species. The daytime average α calculated from the model was ~ 0.126, whilst α calculated from the individual VOCs was ~0.08 demonstrating that the reaction of OH with model-generated intermediates species like PAN can act as a small source of HCHO and should be taken into account.

Using the α determined by the model, the concentration of OH was calculated and compared to the OH measured during the campaign using 1) the model-predicted kOH and modelled HCHO and 2) the measured kOH and measured HCHO from the campaign. OH calculated using measured kOH and HCHO was found to be greater than the OH calculated using the modelled predicted values and greater than the observed OH. This suggests that missing kOH is likely an unmeasured VOC which acts as a source of HCHO and, as such, α is likely greater than calculated by the model. This finding highlight that the unknown VOCs (if not considered) could lead to OH concentrations being over-estimated using the approach outlined.