Report 9: A Modeling Perspective on Cloud Radiative
Forcing
Potter, Gerald L., Julia M. Slingo, Jean-Jacques Morcrette and
Lisa Corsetti
February 1993, 24 pp.
Radiation fields from a perpetual July integration of a T106 version
of the ECMWF operational model are used to identify the most appropriate
way to diagnose cloud radiative forcing in a general circulation model,
for the purposes of intercomparison between models. Differences between
the Methods I and II of Cess and Potter (1987) and a variant method are
addressed. Method I is shown to be the least robust of all methods, due
to the potential uncertainties related to persistent cloudiness, length
of the sampling period and biases in retrieved clear-sky quantities due
to insufficient sampling of the diurnal cycle.
Method II is proposed as an unambigtuous way to produce consistent radiative
diagnostics for intercomparing model results. The impact of the three methods
on the derived sensitivities and cloud feedbacks following an imposed change
in sea surface temperature is discussed. The sensitivity of the results
to horizontal resolution is considered by using the diagnostics from parallel
integrations with a T21 version of the model. (pdf
file)
UCRL-MI-123395