FORCING FUNCTIONS FOR SECOND SET OF INTEGRATIONS

Following are details concerning additional forcing functions to be used in the second set of Climate of the 20th Century model integrations. In some cases, questions are included that are to be addressed at the first C20C Workshop. To elaborate on the data required for the second set of runs, the C20C Organizing Committee recommends five or six forcings:

- SOLAR: This is simply a time series of perturbations to the solar constant, taken from Lean et al., which can be provided by the Hadley Centre (HC). In the HC model, there is additional code to make appropriate adjustments for each spectral band; other model groups may or may not wish to make similar alterations.

VOLCANOES: We will provide the Apr 1995 version of Sato et al.'s dataset, which is available on the WWW. The only manipulation of the data is to degrade the resolution to four latitudinal bands so that the resolution is more consistent throughout the whole period. There is a June 1999 version of the data available as well which can be considered optional.

ANTHROPOGENIC GREENHOUSE GASES: We can provide time series of carbon dioxide, methane, nitrous oxide, CFCs, HFCs and an HCFC. For those participants whose model does not deal with these gases separately, and wish to do this experiment, an equivalent CO2 time series will be provided instead.

SULFATE AEROSOLS (DIRECT EFFECT): In the HC model there is an emissions scenario (from Orn to 1970, GEIA.1B for 1985, and IPCC-SRES from 1990), and aerosol mass mixing ratios are computed with an interactive sulfur chemistry model (which then feed into the radiation scheme).

We can envisage three options for participants:

  1. For those models having the option of interactive sulfur chemistry, they would presumably also use an emissions scenario. We think that the datasets used by the HC are all that are available, so are probably already being used by these groups. Is there a need for us to distribute these data sets?
  2. For models which do not have sulfur chemistry, but can accept aerosols, we could provide a dataset of the anthropogenic aerosol mass mixing ratios from a recent HC model run. This would consist of 19 full global fields [one for each level] for a number of time points. However, this may not be the best available, and other centers may wish to offer an alternative.
  3. For other models, we could provide a dataset of column-integrated aerosol mass per unit area (one global field per month), from the HC model, from which surface albedo perturbations could be computed using the published formula of Charlson et al. We would like to know in advance which option is preferred by each group, so as not to have to produce datasets that no group may require.

SULFATE AEROSOLS (INDIRECT EFFECT): For various reasons, this effect is estimated in the HC model by providing it with adjustments to the cloud albedo, which have been computed from some time-slice experiments that include the interactive sulfur chemistry model and cloud microphysics such that droplet size is dependent on CCNs. However, the HC aerosol experts believe that there would be large inconsistencies, and hence errors, if these perturbations were provided for use in other models. Furthermore, it would also be inappropriate to use a dataset of anthropogenic aerosol mass mixing ratios (mentioned above for the direct effect), since the indirect effect of anthropogenic aerosols is highly dependent on the background concentration of natural aerosols (which would thus then need to be provided for both the 'SST-only' and 'all-forcings' experiments). Hence, we are of the view that this effect can only be included by those models that have options to include both interactive sulfur chemistry and interactive cloud droplet size.

OZONE (STRATOSPHERIC AND TROPOSPHERIC): This is computed from an off-line chemistry model; the HC could provide these data (19 full global fields for every month). However, there is an issue that for (at least) some models, the ozone climatology used in HadISST-only runs may be substantially different from that of the HC model in some regions, and hence the anomalous ozone forcing (and its atmospheric effect) would also be substantially different. One alternative would be to provide the ozone as anomalies from the HC climatology, but this may instead result in unrealistic actual ozone amounts when added to a different climatology. This issue should be resolved by participants in their own preferred way.