Background
The authors write that future projections of climate "have been -- for a given climate model -- derived using a 'standard' set of cloud parameters that produce realistic present-day climate." However, they say "there may exist another set of parameters that produces a similar present-day climate but is more appropriate for the description of climate change," and that "due to the high sensitivity of aerosol forcing (F) to cloud parameters, the climate projection with this set of parameters could be notably different from that obtained from the standard set of parameters, even though the present-day climate is reproduced adequately," which state of affairs suggests that replication of the present-day climate is no assurance that a climate model will accurately portray earth's climate at some future time.

What was done
To get a better idea of the magnitude of uncertainty associated with this conundrum, Haerter et al. used the ECHAM5 atmospheric general circulation model (GCM), which includes parameterizations of direct and first indirect aerosol effects, to see what degree of variability in F results from reasonable uncertainties associated with seven different cloud parameters: the entrainment rate for shallow convection, the entrainment rate for penetrative convection, the cloud mass flux above the non-buoyancy level, the correction to asymmetry parameter for ice clouds, the inhomogeneity parameter for liquid clouds, the inhomogeneity parameter for ice clouds, and the conversion efficiency from cloud water to precipitation.

What was learned
The four researchers report that "the uncertainty due to a single one of these parameters can be as large as 0.5 W/m²," and that "the uncertainty due to combinations of these parameters can reach more than 1 W/m²." As for their significance, they say that "these numbers should be compared with the sulfate aerosol forcing of -1.9 W/m² for the year 2000, obtained using the default values of the parameters."

What it means
With respect to these parametric uncertainties, we apparently do not know the mean sulfate aerosol forcing component of earth's top-of-the-atmosphere radiative budget to within anything better than ± 50%. In addition, Haerter et al. note that structural uncertainties, such as "uncertainties in aerosol sources, representation of aerosols in models, parameterizations that relate aerosols and cloud droplets to simulate the indirect aerosol effect, and in cloud schemes" lead to an overall uncertainty in F of approximately ± 43%, as per the most recent IPCC estimates. In reality, therefore, we probably do not know the current atmosphere's aerosol radiative forcing to anything better that ± 100%, which does not engender confidence in our ability to simulate earth's climate very far into the future with state-of-the-art climate models.