RELATING CIRRUS CLOUD PROPERTIES TO OBSERVED FLUXES - A CRITICAL-ASSESSMENT

The accuracy needed in cirrus cloud scattering and microphysical prope rties is quantified such that the radiative effect on climate can be d etermined. Our ability to compute and observe these properties to with in needed accuracies is assessed, with the greatest attention given to those properties that most affect the fluxes. Model calculations indi cate that computing net longwave fluxes at the surface to within +/- 5 % requires that cloud temperature be known to within as little as +/- 3 K in cold climates for extinction optical depths greater than two. S uch accuracy could be more difficult to obtain than that needed in the values of scattering parameters. For a baseline case (defined in text ), computing net shortwave fluxes at the surface to within +/- 5% requ ires accuracies in cloud ice water content that, when the optical dept h is greater than 1.25, are beyond the accuracies of current measureme nts. Similarly, surface shortwave flux computations require accuracies in the asymmetry parameter that are beyond our current abilities when the optical depth is greater than four. Unless simplifications are di scovered, the scattering properties needed to compute cirrus cloud flu xes cannot be obtained explicitly with existing scattering algorithms because the range of crystal sizes is too great and crystal shapes are too varied to be treated computationally. Thus, bulk cirrus scatterin g properties might be better obtained by inverting cirrus cloud fluxes and radiances. Finally, typical aircraft broadband flux measurements are not sufficiently accurate to provide a convincing validation of ca lculations. In light of these findings we recommend a re-examination o f the methodology used in field programs such as FIRE and suggest a co mplementary approach.