Atmospheric chemistry

No equatorial divide for a cleansing radical

A constraint on the global distribution of the elusive hydroxyl radical takes us a step closer towards understanding the complex, interdependent factors that control the levels of this atmospheric cleanser. See Letter p.219

The atmosphere is cleansed of many air pollutants and some greenhouse gases when these compounds react with the hydroxyl free radical (OH) in the troposphere, the lowermost layer of the atmosphere. Reactions with OH also prevent some substances from destroying the stratospheric ozone layer. Since the OH molecule was first confirmed to exist in the troposphere more than 40 years ago1, a clear depiction of its spatial distribution has been elusive. Roughly speaking, the more OH that is available near pollutant sources, the faster are pollutants removed from the atmosphere, preventing their transport to other atmospheric regions. On page 219 of this issue, Patra et al.2 report that there is little difference in OH abundance in the Northern and Southern hemispheres — in stark contrast to what is currently simulated by global atmospheric-chemistry models.

The OH radical reacts with other species in the atmosphere within 1 second, making its direct measurement a technical feat — one that is not possible on the time and space scales necessary to constrain its global distribution. The global mean concentration of OH can instead be derived by analysing measured abundances of proxy compounds that are emitted to the atmosphere and removed primarily by reaction with OH, provided that the magnitude and spatial distribution of the proxy emissions are accurately known. However, long-standing discrepancies exist between the variability of global mean OH concentrations inferred from proxies and those estimated using computational models3,4.

The best proxy for inferring global OH abundance and variability is methyl chloroform. This anthropogenic compound was once used as a solvent, and depletes ozone in the stratosphere. Emitted mostly in the Northern Hemisphere, methyl chloroform abundances are measured around the world by two monitoring networks. Aircraft flights have also sampled latitudinal variations over the Central Pacific Ocean.

Hemispheric differences in the abundances of methyl chloroform contain signatures of the hemispheric OH ratio — the ratio of annual mean OH concentration in the Northern Hemisphere to that in the Southern Hemisphere. More specifically, the differences reflect the combined influence of the amount and location of the proxy's emissions, its transport between the hemispheres and its temperature-dependent chemical depletion by OH. On the basis of observations of methyl chloroform, previously reported modelling5 inferred hemispheric symmetry in OH concentrations. But this was not conclusive because of the uncertainties associated with methyl chloroform emissions, and because the model did not adequately represent transport across the Intertropical Convergence Zone (ITCZ) — a meteorological barrier to interhemispheric air exchange.

The production of methyl chloroform was banned by the Montreal Protocol, so only residual emissions remain. Because the rate at which it is lost from the atmosphere through reactions with OH now exceeds the emission rate, methyl chloroform can be used to infer OH levels more accurately than ever before3. Indeed, Patra et al. find that hemispheric differences in methyl chloroform abundances are no longer sensitive to uncertainties in the spatial patterns of methyl chloroform emissions.

Using differences between methyl chloroform concentrations measured at sites in the Northern and Southern hemispheres, the authors minimized the combined uncertainty associated with the global mean OH concentration and total methyl chloroform emissions. They also concluded that transport across the ITCZ in their model is accurate because it matches observed distributions of sulphur hexafluoride — an unreactive compound whose sources are better known than those of methyl chloroform and which can therefore be used to infer such transport. Having thus minimized potential sources of uncertainty, the researchers imposed spatially distinct distributions of OH on the model, and demonstrated a strong relationship between their simulated hemispheric OH ratio and the calculated difference in hemispheric abundances of methyl chloroform.

They found that the best match between measured and modelled interhemispheric differences in methyl chloroform levels occurs for roughly equal OH abundances in the two hemispheres. This finding directly conflicts with estimates from current global models of atmospheric chemistry, which consistently simulate higher OH levels in the Northern Hemisphere4. It should be noted, however, that there are open questions associated with those models: they estimate global annual mean OH concentrations that differ by ±25%, and calculate opposing responses to identical changes of anthropogenic emissions4,6.

Although Patra and co-workers provide an invaluable service by pinning down the hemispheric OH ratio, this finding offers little insight into the complex, interdependent processes that shape OH distributions and their temporal evolution. For example, the authors say that overestimates of OH levels in the Northern Hemisphere reflect the tendency of models to calculate higher concentrations of tropospheric ozone — the main OH source — than are actually observed in that region7. But because ozone production in the troposphere requires OH, the calculated high ozone levels could be just another symptom of a common underlying problem: incomplete representation of some atmospheric chemical and physical processes that shape OH and ozone distributions.

Tropospheric ozone generation occurs when ultraviolet radiation splits apart ozone in the presence of water vapour. But OH can also be regenerated when methane and certain other compounds react in the presence of sufficient concentrations of nitrogen oxides — pollutants emitted by cars and smokestacks (Fig. 1; see also ref. 8). The global mean abundance and distribution of OH thus represent the net summation of these photochemical processes over myriad local environments, each of which may be dominated by a different, often poorly constrained, factor. It is changes in these individual factors that determine the evolution of OH.

Figure 1: Sources and sinks of hydroxyl (OH) radicals.

This highly simplified scheme depicts factors that control the abundance and spatial distribution of OH in the troposphere; sources of OH are shown in red, and sinks in blue. The main pathway for OH formation is the absorption of ultraviolet light by tropospheric ozone (O3) and its subsequent break-up in the presence of water vapour. OH radicals may then react with compounds such as methane (CH4), carbon monoxide (CO) and non-methane volatile organic compounds (NMVOCs) to form oxidation products. Some of those products serve as extra sinks for OH before their eventual loss from the atmosphere. Others can react with nitrogen oxides (NOx) in the presence of sunlight (at wavelengths of less than about 420 nanometres) to produce more OH and O3. The relative importance of each source or sink varies spatially and temporally.

Atmospheric-chemistry models are our best tools for estimating the evolution of the atmosphere's self-cleansing capacity, and for evaluating the global impacts of societal choices regarding emissions of air pollutants, greenhouse gases and ozone-depleting substances. But observational constraints with which to test these models are fairly limited. Patra and colleagues' study provides a prime example of how observational constraints can be derived, with immediate ramifications for those working in the field. For instance, the IGAC/SPARC Chemistry-Climate Model Initiative (CCMI) is coordinating a set of simulations using models that serve as workhorses for projecting the evolution of OH in the atmosphere. These simulations should be scrutinized for clues to the key processes determining the parity of hemispheric OH concentrations.

More broadly, Patra and co-workers' study implies that analysing large sets of model simulations, for example those produced through efforts such as the CCMI, can reveal clear relationships between an uncertain model parameter and a directly observable quantity. New observation-derived constraints are sorely needed to work out the complex chemical and physical processes that continuously remove harmful substances from the atmosphere.


  1. 1

    Levy, H. II Science 173, 141–143 (1971).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Patra, P. K. et al. Nature 513, 219–223 (2014).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Montzka, S. A. et al. Science 331, 67–69 (2011).

    ADS  CAS  Article  Google Scholar 

  4. 4

    Naik, V. et al. Atmos. Chem. Phys. 13, 5277–5298 (2013).

    ADS  Article  Google Scholar 

  5. 5

    Krol, M. C. & Lelieveld, J. J. Geophys. Res. 108, 4125 (2003).

    Article  Google Scholar 

  6. 6

    Voulgarakis, A. et al. Atmos. Chem. Phys. 13, 2563–2587 (2013).

    ADS  Article  Google Scholar 

  7. 7

    Young, P. J. et al. Atmos. Chem. Phys. 13, 2063–2090 (2013).

    ADS  Article  Google Scholar 

  8. 8

    Wennberg, P. O. Nature 442, 145–146 (2006).

    ADS  CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Arlene M. Fiore.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Fiore, A. No equatorial divide for a cleansing radical. Nature 513, 176–178 (2014).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing