Constraints and biases in a tropospheric two-box model of OH.

The hydroxyl radical (OH) is the main atmospheric oxidant and the primary sink of the greenhouse gas CH₄. In two recent studies, constraints on the hydroxyl radical (OH) were derived using a tropospheric two-box model of methyl chloroform (MCF) and CH₄. When OH variations as derived in this set-up were propagated to the CH₄ budget, the constraints on OH from MCF still allowed for a wide range of CH₄ emission scenarios. This is important, because global CH₄ emissions are generally considered best constrained by the global lifetime of CH₄, which is determined mainly by OH. Here, we investigate how the use of a tropospheric two-box model in these studies can have affected derived constraints on OH, due to the simplifying assumptions inherent to a two-box model. First, instead of prescribing fixed model parameters for interhemispheric transport, chemical loss rates and loss to the stratosphere, we derive species- and time-dependent quantities from a full 3D transport model simulation. We find significant deviations between the magnitude and time-dependence of the parameters we derive, and the assumptions commonly reported and adopted in literature. Moreover, using output from the 3D model simulations, we investigated differences between the burden seen by the surface measurement network of the National Oceanic and Atmospheric Administration and the true tropospheric burden. Next, we accounted for these biases in a two-box model inversion of MCF and CH₄, to investigate the impact of the biases on OH constraints. We find that the sensitivity of interannual OH anomalies to the biases is modest (1-2%), relative to the significant uncertainties on derived OH (5-8%). However, in an inversion where we implemented all four bias corrections simultaneously, we did find a shift to a positive OH trend over the 1994-2015 period. Moreover, the magnitude of derived global mean OH and by extent that of global CH₄ emissions are affected much more strongly by the bias corrections than their anomalies (∼ 10%). In this way, we identified and quantified direct limitations in the two-box model approach that can possibly be corrected for when a full 3D simulation is used to inform the two-box model. This derivation is, however, an extensive and species-dependent exercise. Therefore, a good alternative would be to move the inversion problem of OH to a 3D model completely. It is crucial to account for the limitations of two-box models in future attempts to constrain the atmospheric oxidative capacity, especially because though MCF and CH₄ behave similarly in large parts of our analysis, it is not obvious that this should be the case for alternative tracers that potentially constrain OH, other than MCF. [ABSTRACT FROM AUTHOR]