On the information content of the thermal infrared cooling rate profile from satellite instrument measurements

Abstract

[1] This work investigates how remote sensing of the quantities required to calculate clear-sky cooling rate profiles propagates into cooling rate profile knowledge. The formulation of a cooling rate profile error budget is presented for clear-sky scenes given temperature, water vapor, and ozone profile uncertainty. Using linear propagation of error analysis, an expression for the cooling rate profile covariance matrix is given. Some of the features of the cooling rate covariance matrix are discussed, and it is found that nonzero error correlations in the temperature, water vapor, and ozone retrieval profiles must be considered to produce an unbiased estimate of cooling rate profile variance and the covariance structure. To that end, the exclusion of the details of this error correlation leads to an underestimation of the cooling rate profile uncertainty. This work then examines the assumptions made in the course of deriving the expression for the cooling rate covariance matrix by using ERA-40 Reanalysis data. It is established that the assumptions of linear error propagation and Gaussian statistics are generally tenable. Next, the information content of thermal infrared spectra with respect to clear-sky cooling rate profiles is investigated. Several formerly-and currentlyoperational spectrometers are compared with different spectral coverage, resolution, signal-to-noise ratio. Among these, IASI is found to have the ability to provide the greatest amount of information on the cooling rate profile. Also, it may be scientifically useful to develop far-infrared missions in terms of cooling rate profile analysis.

ENSO regulation of far‐ and mid‐infrared contributions to clear‐sky OLR

“…Spectral observations in the FIR (mean and variance), especially in moist low-latitude regions, may yield additional insight on important mixing processes and impacts on the radiation budget that operate in climate models. Pierrehumbert and Yang [1993] showed that tropical mixing is likely dominated by cross-isentropic transport from diabatic heating of convective latent heat release and radiative cooling in clear [e.g., Allan et al, 1999;Feldman et al, 2008] and cloudy skies [e.g., Mather et al, 2007]. GCMs are known to artificially diffuse small-scale structures that are generated by mixing processes [e.g., Skamarock, 2004] and in turn may inadvertently enhance meridional transport through the tropical mixing barrier [Webster and Holton, 1982].…”

Effect of correlated observation error on parameters, predictions, and uncertainty

“…When calibration observations are derived from multiple direct measurements, the observation error variances and covariances can be estimated by propagating the measurement error (Figure ) [e.g., Sherman , ; Tellinghuisen , ; Feldman et al ., ]. As discussed in section 1, derived observations commonly are used to calibrate hydrologic models.…”

Remote sensing of tropical tropopause layer radiation balance using A‐train measurements

“…The current algorithm used to create CloudSat 2B‐FLXHR products does not include error estimates on fluxes and heating rate products, though this can be accomplished through formal error propagation analysis [ Taylor and Kuyatt , 1994; Feldman et al , 2008]. Assuming that the variables relevant to heating rate calculations can be statistically modeled as Gaussian, the uncertainty in the cooling rate profile is given by the following: where ( x 1 , … x i , … x j , …, x n ) represent all of the atmospheric state inputs that are relevant to cooling rate profile calculations at each level, θ′( z ) refers to the broadband cooling rate at height z , and cov refers to the covariance function.…”

“…[14] The current algorithm used to create CloudSat 2B-FLXHR products does not include error estimates on fluxes and heating rate products, though this can be accomplished through formal error propagation analysis [Taylor and Kuyatt, 1994;Feldman et al, 2008]. Assuming that the variables relevant to heating rate calculations can be statistically modeled as Gaussian, the uncertainty in the cooling rate profile is given by the following:…”