Background
Tropospheric ozone is an important greenhouse gas that absorbs both longwave (terrestrial) and shortwave (solar) radiation.  Produced by the photochemical oxidation of CO and hydrocarbons in the presence of NO and NO2, large increases in tropospheric ozone have been observed over the past century, as anthropogenic emissions of its precursors have risen dramatically.

The resultant effect on climate since preindustrial times has typically been estimated to be a radiative forcing between 0.3 and 0.5 watts per square meter.  However, as Mickley and Jacob note, there is "considerable uncertainty about ozone levels in preindustrial times," and current models "overestimate systematically the late nineteenth and early twentieth century observations."  As a result, climate models may currently be significantly underestimating the contribution of tropospheric ozone to the global warming observed since the end of the Little Ice Age.

What was done
The authors used a global three-dimensional model of tropospheric chemistry to explore "to what extent uncertainties in natural sources for the nineteenth century can accommodate the low levels of ozone observed."  Furthermore, they examined the consequences of such uncertainties on model assessments of the radiative forcing increase from tropospheric ozone since preindustrial times.

What was learned
Model simulations revealed that uncertainties associated with natural emissions of NOx and hydrocarbons led to decreases in preindustrial ozone concentrations of 10-20 parts per billion relative to the values state-of-the-art models use in making calculations of tropospheric ozone radiative forcing.  Test simulations using these new estimates yielded "a global mean radiative forcing from ozone added to the atmosphere since preindustrial times of 0.72-0.80 watts per square meter, well above the range of forcings (0.3-0.5 watts per square meter) obtained by standard models."

What it means
In considering their results, the authors acknowledge there are uncertainties in their estimates that could account for the higher global radiative forcing produced by the model they used.  However, they note that "the test simulations represent our best reconstructions of the global preindustrial ozone fields constrained by the 1870-1910 surface air observations."  Thus, they conclude that "tropospheric ozone may have contributed a much larger fraction of total greenhouse forcing since preindustrial times than is generally assumed."  Such a result is very important, for if correct, it amounts to about half of the estimated increase in CO2-induced radiative forcing during the same time period, which would suggest that estimates of CO2-induced radiative forcing since preindustrial times are considerably overstated.