| Abstract: |
The Global Precipitation Measurement (GPM) mission uses a constellation of microwave radiometers on various satellite platforms to measure global precipitation with high temporal and spatial resolution. To obtain accurate and consistent measurements from the constellation, the radiometers are intercalibrated to a common reference standard, the GPM Microwave Imager (GMI) on the GPM Core Observatory, launched in February 2014. This constellation recently expanded to include those radiometers in operation prior to the launch of GPM, dating back to the launch of the Tropical Rainfall Measuring Mission (TRMM) in November 1997. Including these radiometers into the GPM constellation extends the data record over many years for studying long-term precipitation trends.
Three of the TRMM era radiometers that were incorporated into the GPM constellation are the Advanced Microwave Sounding Units (AMSU-B), flying on-board the NOAA15, NOAA16, and NOAA17 platforms. These radiometers are important to the GPM constellation as they provide a significant portion of the constellation observations during the TRMM era. Unfortunately, the AMSU-B instruments are known to have considerable calibration issues and require detailed analysis and quality control to ensure accurate precipitation estimates. The GPM Intercalibration Working Group (XCAL) recently analyzed the AMSU-B datasets and delivered intercalibration constants to adjust the AMSU-B calibration to GMI for the GPM Version 5 (V05) data release. During XCAL's initial analysis of the AMSU-B instruments, there were several calibration issues noted but not all were corrected in the V05 dataset. NOAA15 and NOAA16 both show significant time-dependent and scan-dependent calibration issues while NOAA17 shows relative stability similar to what is observed with the higher-quality Microwave Humidity Sounder (MHS) instruments on the NOAA18, NOAA19, Metop-A, and Metop-B platforms. In the AMSU-B V05 dataset, a constant offset for each channel of each instrument is applied to adjust the calibration to GMI. Data are set to missing where it was determined that the calibration had degraded to a point where it would negatively impact retrievals. This is done for the NOAA16 183 GHz channels from January 2008 onward, the NOAA15 183 GHz channels from January 2009 onward, and the NOAA 15 183±3 GHz channel for the entire mission. There is no adjustment in V05 for time-dependent or scan-dependent biases, as it was not yet determined whether these biases have an impact on the precipitation retrievals.
This presentation will present the results of the AMSU-B intercalibration analysis that is included in the GPM V05 dataset and discuss the additional corrections that will be looked at for the next data version release (V06). In addition, the impact of the AMSU-B calibration on the precipitation retrievals will be discussed and the quality of the AMSU-B retrievals analyzed by performing comparisons with MHS. As a result of this analysis, calibration issues with AMSU-B that negatively impact the precipitation retrieval can be identified and corrected in the next version.
|
