Articles | Volume 12, issue 19
Atmos. Chem. Phys., 12, 9097–9111, 2012
Atmos. Chem. Phys., 12, 9097–9111, 2012

Research article 04 Oct 2012

Research article | 04 Oct 2012

Radiative impacts of cloud heterogeneity and overlap in an atmospheric General Circulation Model

L. Oreopoulos1, D. Lee1,2,3, Y. C. Sud1, and M. J. Suarez1 L. Oreopoulos et al.
  • 1NASA Goddard Space Flight Center, Greenbelt, MD, USA
  • 2University Space Research Association, Columbia, MD, USA
  • 3Seoul National University, Seoul, South Korea

Abstract. The radiative impacts of horizontal heterogeneity of layer cloud condensate, and vertical overlap of both condensate and cloud fraction are examined with the aid of a new radiation package operating in the GEOS-5 Atmospheric General Circulation Model. The impacts are examined in terms of diagnostic top-of-the atmosphere shortwave (SW) and longwave (LW) cloud radiative effect (CRE) calculations for a range of assumptions and overlap parameter specifications. The investigation is conducted for two distinct cloud schemes, one that comes with the standard GEOS-5 distribution, and another used experimentally for its enhanced cloud microphysical capabilities. Both schemes are coupled to a cloud generator allowing arbitrary cloud overlap specification. Results show that cloud overlap radiative impacts are significantly stronger in the operational cloud scheme where a change of cloud fraction overlap from maximum-random to generalized results in global changes of SW and LW CRE of ~4 Wm−2, and zonal changes of up to ~10 Wm−2. This is an outcome of fewer occurrences (compared to the other scheme) of large layer cloud fractions and fewer multi-layer situations where large numbers of atmospheric layers are simultaneously cloudy, both conditions that make overlap details more important. The impact of the specifics of condensate distribution overlap on CRE is much weaker. Once generalized overlap is adopted, both cloud schemes are only modestly sensitive to the exact values of the overlap parameters. When one of the CRE components is overestimated and the other underestimated, both cannot be driven simoultaneously towards observed values by adjustments to cloud condensate heterogeneity and overlap specifications alone.

Final-revised paper