Using CRTM to Validate ATMS Data
CICS-MD Scientist QuanhuaLiu and JCSDA’s Sid Boukabara recently published an article documenting the application of the Community Radiative Transfer Model (CRTM) for Advanced Technology Microwave Sounder (ATMS) data validation in the journal Remote Sensing of the Environment. They analyzed the bias between ATMS measurements and the CRTM simulations using the ECMWF analysis. The bias is mainly caused by the uncertainties in the model inputs, the error in cloud identification, and uncertainties in the surface emissivity model. In order to avoid the complexity of microwave land surface emissivity, analyses were carried out over oceans. Clouds are a big issue in the comparison because there are large uncertainties in cloud variables. Traditionally, clouds are screened out by using cloud liquid water path, estimated from microwave window channels at two low frequencies (Weng and Grody, 1994). The horizontal resolution of the ATMS window channels 1 and 2 is very poor and quite different from the spatial resolution for other channels. The various spatial resolutions among the ATMS channels are challenging to the cloud screen method. In the paper, they use the uniformity to exclude cloudy pixels. ATMS channel 3 at 50.3 GHz, with fields of view averaged to 3 × 3, is used for the uniformity test. A threshold of 0.7K is used. The uniformity test result at channel 3 is used for all other channels. As shown in Figure 1, the large tail (red line) due to clouds on the right side is reduced (green line). Over oceans, clouds contribute more radiation to the ATMS measurements than oceans. There are still tails on the right side in green curves in which the uniformity test at channel 3 is applied. When they further limit data within 50% in the histogram of channel 3, they found (see black line) that more cloudy pixels are excluded. The uniformity test left 30% of total data points for use. The second condition (50% histogram) further reduces the data by 10%.
Figure 1. Histograms of the ATMS observations minus the CRTM simulations over oceans. The red line uses all data. The green line is for the data which passed the uniformity test of channel 3. The black line is for the data which passed the uniformity test and within central 50% of the histogram of channel 3.
Liu, Q., and S. Boukabara, 2014: Community Radiation Transfer Model (CRTM) Applications in Supporting the Suomi National Polar-Orbiting Partnership (SNPP) Mission validation and verification, Remote Sen. Environ., 140, 744–754.