Fluctuation Statistics of Outgoing Longwave Radiation in a General Circulation Model and in Satellite Data

View More View Less
  • 1 Atmospheric Sciences Division, NASA Langley Research Center, Hampton, Virginia
  • | 2 Aero-Space Technologies Division, PRC Kentron, Hampton, Virginia
© Get Permissions
Full access

Abstract

The fluctuation statistics of Northern Hemisphere winter outgoing longwave radiation (OLR) computed by a general circulation model (GCM) are compared with those obtained from Nimbus 7 and NOAA AVHRR satellite observations. The model cloud areas, heights, and infrared emissivities change in response to other variables. The effect of the GCM clouds on the OLR is diagnosed by computing separate statistics for the GCM clear sky OLR. This permits us to isolate the effect of clouds on the fluctuation of radiation.

The GCM OLR has more variance than does the satellite, but the GCM clear sky OLR has less variance than the satellite OLR. The autocorrelation of GCM OLR at fixed points (Eulerian correlation) decays faster than the observed OLR; much of this decay is caused by fluctuations in humidity and temperature, as well as cloudiness. The geographical variation of the 24-hour autocorrelations of OLR in the GCM, the GCM clear sky, and the satellites are similar; they display minima in the zones of maximum variance of the band-pass (2.5- to 6-day periods) and high-pass (1- to 2-day) filtered 500-mb geopotential height. The increase in autocorrelation with averaging area, the spatial correlation, and the space-time advection of correlation are similar in the GCM and satellites. The GCM cloud routines produce a negatively skewed radiation field, which is consistent with the observations. The satellite observations suggest that tropical cloud regimes can decrease the persistence of OLR over some regions, but can be associated with high persistence over others.

Abstract

The fluctuation statistics of Northern Hemisphere winter outgoing longwave radiation (OLR) computed by a general circulation model (GCM) are compared with those obtained from Nimbus 7 and NOAA AVHRR satellite observations. The model cloud areas, heights, and infrared emissivities change in response to other variables. The effect of the GCM clouds on the OLR is diagnosed by computing separate statistics for the GCM clear sky OLR. This permits us to isolate the effect of clouds on the fluctuation of radiation.

The GCM OLR has more variance than does the satellite, but the GCM clear sky OLR has less variance than the satellite OLR. The autocorrelation of GCM OLR at fixed points (Eulerian correlation) decays faster than the observed OLR; much of this decay is caused by fluctuations in humidity and temperature, as well as cloudiness. The geographical variation of the 24-hour autocorrelations of OLR in the GCM, the GCM clear sky, and the satellites are similar; they display minima in the zones of maximum variance of the band-pass (2.5- to 6-day periods) and high-pass (1- to 2-day) filtered 500-mb geopotential height. The increase in autocorrelation with averaging area, the spatial correlation, and the space-time advection of correlation are similar in the GCM and satellites. The GCM cloud routines produce a negatively skewed radiation field, which is consistent with the observations. The satellite observations suggest that tropical cloud regimes can decrease the persistence of OLR over some regions, but can be associated with high persistence over others.

Save