CRTM Upgrades and Applications for GOES-R Program

2017 ANNUAL REPORT

Background

In preparing for the incoming launch of the GOES-R, we will update the current CRTM to accommodate for both the scientific and operational needs for the future applications of GOES-R ABI data. Work related to GOES-R algorithm developments includes updating the CRTM capability for simulating AHI and ABI visible channel reflectance for aerosols and clouds; updating and refining look-up-tables (LUT) for clouds, aerosols and precipitations; and updating surface emissivity datasets for surface-sensitive AHI/ABI channels. In order to enhance the calibration and validation (CalVal) science activities for GOES-R, CRTM will also be extended from the current scalar radiative transfer model to a new vector radiative transfer model. The vector CRTM can provide solution of the full Stokes vector with the coupled information. In addition, we will continue working on improving the current emissivity models of polarized surfaces (e.g., ocean and land). An implementation of the new surface emissivity models to the vector CRTM will also play a key role to the instrument CalVal process. Work related to GOES-R Algorithm Working Group (AWG) applications in numerical weather prediction (NWP) include cloud detection, bias estimation, and data thinning, which are works that require careful investigation and be completed before any AHI/ABI data can be effectively assimilated in NWP models.

Accomplishments

Starting 2014, the new generation of Japanese geostationary meteorological satellite carries an Advanced Himawari Imager (AHI) to provide the observations of visible, near-infrared and infrared with much improved spatial and temporal resolutions. For applications of the AHI measurements in numerical weather prediction (NWP) data assimilation systems, the biases of the AHI brightness temperatures at channels 7-16 from the model simulations are firstly characterized and evaluated using both Community Radiative Transfer Model (CRTM) and Radiative Transfer for TIROS Operational Vertical Sounder (RTTOV). It is found that AHI biases under clear atmospheres are independent of satellite zenith angle except for channel 7. The biases of three water vapor channels increase with scene brightness temperatures and the rest of infrared channels are nearly constant except at high brightness temperatures. The AHI biases at all the infrared channels over ocean and land are less than 0.6 K and 1.2 K, respectively. The bias differences between CRTM and RTTOV are small except for the two upper tropospheric water vapor channels 8-9 and the low tropospheric carbon dioxide channel 16. Since the inputs used for simulations are the same for CRTM and RTTOV, the differential biases at the water vapor channels may be associated with the subtle difference in forward models. Figure 1 shows the SRFs and the center frequencies of AHI channel 16 with and without using the updated SRFs. The center frequency of channel 6 shifted from a larger value (753.37 cm^-1)to a smaller value (754.13 cm^-1).

Figure 1. (a) Spectral response functions (curves) and the center frequencies of AHI channel 16 with (blue) and without (red) using the updated SRFs, as well as a typical IASI brightness temperatures simulated with CRTM (grey). (b) Biases of AHI channel 16 with model simulations generated by CRTM with old SRF (black), CRTM with new SRF (blue), RTTOV with new SRF (red), CRTM with old SRF and RTTOV emissivity (grey), and CRTM with the new SRF and RTTOV emissivity (cyan).

Because the spectral radiances in channel 16 have strong slope, the SRF shift can cause a large difference between CRTM and RTTOV. In fact,the bias of AHI channel 16 calculated by O-B^CRTM with the updated AHI SRFs reduces to -0.52 K, which is comparable to that of RTTOV. In other words, a 0.24cm^-1 center frequency and SRF shift doubled the bias magnitude of channel 16.

Planned work        

• Compare CRTM and RTTOV simulations and performance over various surface conditions and document the findings on the discrepancy between the two models
• Characterize the ABI O-B bias with respect to scan angles and under different surface types and find out further improvements needed for CRTM
• Develop a set of cloud mask tests that can be used for cloud detection of ABI data assimilation for NWP
• Propose a resolution-dependent “optimal” AHI/ABI data thinning scheme for AHI/ABI data assimilation in NWP models at different horizontal and vertical resolutions

Publications           

Zou, X., X. Zhuge and F. Weng, 2016: Characterization of bias of Advanced Himawari Imager infrared observations from NWP background simulations using CRTM and RTTOV. J. Atmos. Oceanic Technol.,33, 2553-2567. doi: 10.1175/JTECH-D-16-0105.1.

Presentations       

“Assimilation of AHI Infrared Radiance Measurements for Improved Typhoon Track and Intensity Forecasts Using HWRF” by Zou, Qin, Zhuge, and Weng at the Fifth AMS Symposium on the Joint Center for Satellite Data Assimilation.