Comparison with LBL simulations

Comparison with LBL simulations

RTTOV-LBL comparisons for VIS/IR coefficient files

The plots linked below for each coefficient file show statistics of comparisons of RTTOV radiances with those computed from the line-by-line (LBL) calculations. The comparisons are in terms of brightness temperature (BT) for channels with wavelengths greater than 3 microns and in terms of reflectance for channels with wavelengths less than 3 microns. Simulations are run over the diverse 83 profile set which is used to train RTTOV.

There are two types of plots: the first kind shows comparisons of RTTOV radiances with radiances calculated using the RTTOV integration of the radiative transfer equation (RTE) applied to the LBL channel-integrated optical depths. This gives an indication of the errors resulting from the optical depth regression.

The second kind of plot shows comparisons of RTTOV radiances with channel-integrated radiances. For non-hyperspectral sensors the channel-integrated radiances are calculated using the RTTOV integration of the RTE applied to the LBLRTM transmittances at a spectral resolution of 0.25cm-1 wavenumbers for all profiles and zenith angles used in the coefficient training. The resulting radiances are integrated over the channel spectral responses. This gives an indication of the total errors from both the optical depth regression and the calculation of radiances from polychromatic optical depths. These plots are not generated for channels at wavelengths below 3µm.

For hyperspectral IR sounders this second type of plot uses radiances at a spectral resolution of 0.001cm-1 wavenumbers integrated over the spectral response function for the training profiles at nadir only.

For channels at wavelengths above 3µm, the simulations are run for all zenith angles used in the coefficient training (except for the hyperspectral sounder plots comparing channel-integrated LBLRTM radiances, as noted above) and include contributions from atmospheric emission and emission from a surface with unit emissivity located at the bottom level of the coefficient pressure profile (1050hPa for 54L coefficients, 1100hPa for 101L coefficients).

For channels at wavelengths below 3µm, the simulations are run for a range of satellite and solar zenith angles with a relative azimuth of 180 degrees. The simulations include contributions from atmospheric Rayleigh scattering and surface reflection assuming a surface BRDF of 0.3/pi located at the bottom level of the coefficient pressure profile (1050hPa for 54L coefficients, 1100hPa for 101L coefficients).

The plots show the average (mean), standard deviation (in some cases), RMS and maximum absolute difference between the RTTOV and LBL radiances calculated over the diverse profile set.

Histograms: for a small number of files (currently v9 predictor IASI and SO2 IASI), histograms of the LBLRTM-RTTOV differences have also been plotted for a subset of channels. These show the differences between RTTOV and the BTs calculated from the channel-integrated LBLRTM transmittances. Gaussian distributions with the mean and standard deviation of the differences are also plotted for comparison.



Hyperspectral IR sounders

v7 predictor 54L coefficient files
IR-only, variable O3
v8 predictor 54L coefficient files
IR-only, variable O3+CO2
v7 predictor 101L coefficient files
IR-only, variable O3
v8 predictor 101L coefficient files
IR-only, variable O3+CO2
v9 predictor 101L coefficient files
VIS+IR, all gases



Non-hyperspectral visible/IR sensors


v7 predictor 54L coefficient files
IR-only, variable O3
v8 predictor 54L coefficient files
IR-only, variable O3+CO2
v9 predictor 54L coefficient files
VIS+IR, variable O3+CO2 or O3

Comparisons against LBL calculations for MW v7 predictor coef files


The files linked below show the mean and RMS difference between brightness temperatures calculated using the LBL channel-integrated optical depths and the RTTOV predicted optical depths. The brightness temperatures are computed for nadir view (zenith angle 0 degrees) with unit surface emissivity and comprise upwelling atmospheric emission plus the surface emission term. The statistics are calculated for the 83 diverse profile set used to train RTTOV and give an indication of the error due to the optical depth regression expressed in terms of TOA brightness temperature.