Seasonality and knowledge gaps: Total hydroxyl radical (OH) reactivity measurements in the boreal forest

After a long journey to get them published, the results of our hydroxyl radical (OH) reactivity measurements in the boreal forest in 2016 are now available in Atmospheric Chemistry and Physics.

The final manuscript has been substantially modified from its discussion version, mostly to improve its readability and remove unnecessary parts. The most consequential changes did not severely impact our conclusions, though. Here I give a summary of the findings from our study that includes measurements at the SMEAR II boreal forest station in Hyytiälä, Finland, from April to July 2016, as well as modelling results from our colleagues at the University of Helsinki.

• We found large total OH reactivity values occurred when the soil was thawing after snow melted (late April). These reactivity peaks were even higher than the high total OH reactivity values in summer (July).

• Comparing the total OH reactivity measured with the OH reactivity calculated from the known chemical composition at the site, it is possible to calculate the amount of missing (or unexplained) reactivity. In this study, the missing reactivity fraction remained high (more than 50% most of the time) despite including more compounds than in previous studies at the same site (Sinha et al., 2010; Nölscher et al., 2012). Note that this comparison is not as straightforward as it might seem due to uncertainties on the total OH reactivity measurements on one hand and the fact that the availability of data to calculate the known OH reactivity at the site can vary a lot (e.g. due to instrumental breaks).

• Modelling oxidation products that were not directly measured to include in the calculated OH reactivity did not reduce the missing fraction significantly. However, it should be noted that the model uses the atmospheric chemistry from the Master Chemical Mechanism contains only three monoterpenes and one sesquiterpene, even though more where measured at the site. From these additional terpenes, no oxidation products were modeled, even though there must be some and they contribute certainly to the measured OH reactivity. Also some of the reaction rates used for the calculations are only estimated because experimental values have not been measured.

In conclusion our study demonstrates that there are unknown primary emissions that are probably non-terpene compounds and possibly from sources other than trees (e.g. soil). In addition, the direct measurements of oxidation products would be valuable to investigate their contribution to the total OH reactivity (the precursors measured such as terpenes are very reactive, therefore large amounts of reaction products could be present).