| Paper: | TH-A1.10 |
| Session: | Applications of Radiometry I |
| Time: | Thursday, March 29, 09:00 - 10:20 |
| Presentation: |
Poster
|
| Topic: |
Theory, physical principles and electromagnetic models: |
| Title: |
Including Atmospheric Polarization in Community Radiative Transfer Model (CRTM) |
| Authors: |
Quanhua (Mark) Liu; NOAA/NESDIS Center for Satellite Applications and Research | | |
| | XingMing Liang; ERT Corporation | | |
| | Shouguo Ding; ERT Corporation | | |
| | lin lin; University of Maryland | | |
| Abstract: |
The Community Radiative Transfer Model (CRTM), developed at the Joint Center for Satellite Data Assimilation in the United States, operationally supports satellite radiance assimilation for weather forecast, sensor data verifications, and retrievals of satellite products.
The CRTM is used in radiance monitoring system. By analyzing the statistical difference between microwave radiance observations (e.g. Advanced Technology Microwave Sounder (ATMS) ) and the CRTM simulations, we can monitor the instrumental stability and estimate the absolute calibration accuracy. The CRTM plays a critical role in the NOAA Microwave Integrated Retrieval System (MiRS) for the generation of atmospheric profiles of temperature and water vapor, cloud liquid water, ice water content, rainfall rate, snow cover and snow water equivalent, snow fall rate, surface temperature and microwave emissivity, and sea ice concentration.
CRTM is a channel-based operational radiative transfer model. For a given channel and an atmospheric layer, an effective optical depth is computed to represent the channel transmittance. The CRTM is a scalar radiative transfer model except for the polarization in ocean surface emissivity model. However, neglecting atmospheric polarization under precipitation conditions can result in the simulation error of few Kelvin which can be critical for cloud radiance assimilation and retrieval. In this presentation, we will show the polarization effect. We will also show the effect of ATMS instrument spectral response data. It is worthy to be mentioned that the central frequencies and band widths in a specification document is slightly different from actual measurements in calibration data book. The real spectral response function (SRF) data are different from the rectangular assumption and the simulation difference between using the real and rectangular SRFs can be in the order of 0.1K and 0.2K.
|
