Asynoptic Sampling Considerations for Wide-Field-of-View Measurements of Outgoing Radiation. Part II: Diurnal and Random Space-Time Variability

View More View Less
  • 1 Department of Astrophysical, Planetary and Atmospheric Sciences, University of Colorado, Boulder, Colorado
© Get Permissions
Full access

Abstract

Two classes of tropical cloud variability: (i) random small-scale fluctuations and (ii) diurnal variations, are investigated with regard to deriving fields of emitted radiation from wide-field-of-view (WFOV) measurements of outgoing radiance made aboard polar orbiting satellites. Irregular cloud variability is represented in terms of a stochastic space-time process defined by prescribed spatial and temporal correlation scales and confined to an envelope typical of tropical convective centers. Diurnal cloud variability is prescribed in terms of a propagating solar waveform which likewise is confined to a horizontal envelope. For both classes of convective behavior, the evolving radiation field is sampled asynoptically, deconvolved, and compared with the true variability. A variety of diagnostics is examined, including space-time power spectra, instantaneous synoptic behavior, and time-mean fields. Sensitivity to spatial and temporal scales of the convective pattern, (correlation scales in the case of random variability) is evaluated for both classes of behavior.

For realistic convective scales, the retrieved behavior is aliased by unresolved variability. Of the two classes of convective behavior, diurnal variations pose the most serious challenge to the sampling, because they introduce variance well removed from the Nyquist limits of asynoptic observations. Moreover, while unresolved behavior may be eliminated by time averaging in the case of random variability, diurnal variations alias to steady components whose influence remains even in time-mean fields. These aliases would be expected to seriously contaminate monthly- and seasonal-mean behavior over tropical landmasses, where diurnal viability is large.

Contemporaneous WFOV measurements from several satellites orbiting the globe (e.g., ERBE) may hold the solution to the problem. The expanded information content, represented by the combined data, should capture most if not all of the large-scale variability unresolved by single satellite sampling. An added and unique advantage of WFOV observations is that mesoscale convective complexes, which are predominant in tropical convection but which may only be resolved with a geostationary platform, are automatically removed–without aliasing, leaving behind the large-scale filtered field which is resolvable and of primary interest for many applications. Extending the resolution of asynoptic measurements with multiple satellite data will require that sampling asymmetries, inherent to the combined data ensemble, be explicitly accounted for. With the majority of cloud brightness variability captured by the combined sampling, this would place within grasp faithful recovery of, not only the large-scale time-mean field, but the complete synoptic structure and evolution as well.

Abstract

Two classes of tropical cloud variability: (i) random small-scale fluctuations and (ii) diurnal variations, are investigated with regard to deriving fields of emitted radiation from wide-field-of-view (WFOV) measurements of outgoing radiance made aboard polar orbiting satellites. Irregular cloud variability is represented in terms of a stochastic space-time process defined by prescribed spatial and temporal correlation scales and confined to an envelope typical of tropical convective centers. Diurnal cloud variability is prescribed in terms of a propagating solar waveform which likewise is confined to a horizontal envelope. For both classes of convective behavior, the evolving radiation field is sampled asynoptically, deconvolved, and compared with the true variability. A variety of diagnostics is examined, including space-time power spectra, instantaneous synoptic behavior, and time-mean fields. Sensitivity to spatial and temporal scales of the convective pattern, (correlation scales in the case of random variability) is evaluated for both classes of behavior.

For realistic convective scales, the retrieved behavior is aliased by unresolved variability. Of the two classes of convective behavior, diurnal variations pose the most serious challenge to the sampling, because they introduce variance well removed from the Nyquist limits of asynoptic observations. Moreover, while unresolved behavior may be eliminated by time averaging in the case of random variability, diurnal variations alias to steady components whose influence remains even in time-mean fields. These aliases would be expected to seriously contaminate monthly- and seasonal-mean behavior over tropical landmasses, where diurnal viability is large.

Contemporaneous WFOV measurements from several satellites orbiting the globe (e.g., ERBE) may hold the solution to the problem. The expanded information content, represented by the combined data, should capture most if not all of the large-scale variability unresolved by single satellite sampling. An added and unique advantage of WFOV observations is that mesoscale convective complexes, which are predominant in tropical convection but which may only be resolved with a geostationary platform, are automatically removed–without aliasing, leaving behind the large-scale filtered field which is resolvable and of primary interest for many applications. Extending the resolution of asynoptic measurements with multiple satellite data will require that sampling asymmetries, inherent to the combined data ensemble, be explicitly accounted for. With the majority of cloud brightness variability captured by the combined sampling, this would place within grasp faithful recovery of, not only the large-scale time-mean field, but the complete synoptic structure and evolution as well.

Save