Frequent Asked Questions:
1. How to apply the GSICS GEO-LEO correction?
The correction coefficients (offset and slope) at the GSICS GEO-LEO correction products
in netCDF are derived from the linear relation between the monitored GEO radiance and the reference LEO radiance at the collocated scenes. Here is the equation to apply the GEO-LEO correction:
GSICS_Corrected_Radiance = Monitored_GEO_Radiance/Slope - Offset/Slope (Equation 1)
Here Slope and Offset are the correction coefficients recorded in the netCDF files of GSICS GEO-LEO correction products.
2. What is FOV and ENV areas?
In NOAA GOES vs. AIRS inter-calibration ATBD, on section 2.c.iii.v0.1, there are definitions of target area (FOV) and environment area (ENV). The target area is defined as the nominal LEO FOV centered at the collocated GEO pixel. The environment is defined by the GEO pixels within 3x radius of the target area from the centre of each LEO FOV. Note that both areas are centered at the GEO pixel which is closest to the LEO pixel. For GOES, the ENV area has the the 17x9 GEO pixels in size due to the E-W oversampling.
3. How to use the radiance-based correction algorithm for brightness temperature(Tb) correction?
The GSICS GEO-LEO correction is based on spectral radiance. Here are the procedure to convert to brightness temperature(Tb) correction:
- Step 1: Convert GEO Tb to spectral radiance. There are two ways to get the spectral radiance: (1) Convolve the the spectral distribution of Planck function and spectral response function, Or (2) First covert the Tb to effective brightness temperature (Teff) using the band correction. Then get the spectral radiance with Planck function at Teff and central wave-number values.
- Step 2: Apply the GSICS correction algorithm (Equation 1) to get the GSICS_Corrected_Radiance
- Step 3: Convert the GSICS_Corrected_Radiance to GSICS_Corrected_Tb (1) Use the inverse Planck function to get the effective brightness temperature (GSICS_Correct_Teff) (2) Convert the GSICS_Correct_Teff to GSICS_Correct_Tb with band correction
GOES band correction information can be found at GOES GVAR conversion webpage
4. What is the standard scene and homogeneous scene Tb biases?
The standard scene, which serves as an independent verification of the correction product, should be calculated for each channel a priori using a radiative transfer model based on a standard atmospheric atmospheric profile and surface conditions. However, we have not finished the CRTM model yet (should be coming soon) for all the GOES IR channels. Currently we randomly picked up one month to calculate the mean GOES Tb values as temporarily substitute and shall be replaced with the simulated result very soon.
Standard scene Tb bias (std_tb_bias) = geo_ref_tb - std_scene_tb
where std_scene_tb is the standard scene temperature and geo_ref_tb is GSICS corrected GEO Tb for the standard scene temperature.
Homogeneous scene is identified if the standard deviation of radiance/mean radiance for the array of 3x3 pixels (both GEO and LEO) centered at collocated scenes <0.05. Homogeneous scene Tb bias is defined as the daily mean GEO-LEO Tb difference of the collocation scenes which pass the homogeneous screening.
hmg_tb_bias = mean(GEO_hmg_tb - LEO_hmg_tb)
where GEO_hmg_tb and LEO_hmg_tb are the Tb values for the GEO and LEO homogeneous scenes, respectively.
As Tb bias is a function of scene temperature, the Tb can vary may vary with the spectra distribution of the collocated homogeneous scenes. Unlikely GEO-LEO inter-calibration for Imager which has hundreds to thousands homogeneous scenes very day, we have much less homogeneous scenes for Sounder GEO-LEO inter-calibrations. We are considering to take out this monitoring variables from the product variable list once the standard scene temperature is generated from the RTM models.
Both the stand scene and homogeneous scene Tb values and associated uncertainty (standard errors) are for the purpose of the product quality monitoring. Currently. we don't recommend using these variables for a simplied bias/offset correction.