Abstract
This paper focuses on the development and validation of retrieval models of water vapor (WV) and wet tropospheric path delay (PD) for the calibration microwave radiometer (CMR) on board the first ocean dynamic environment satellite of China—Haiyang-2A (HY-2A). The reference data are those of Jason-1 microwave radiometer (JMR) and Jason-2 advanced microwave radiometer (AMR). The crossover points of the HY-2A and Jason-1/2 satellite tracks are extracted, and the data pairs at these points are divided into fitting and validation datasets. The retrieval models of WV and PD are built for the CMR using the fitting dataset and genetic algorithm, and validated using the validation dataset. The validation shows that the results of the retrieval models are consistent with the JMR and AMR data, and the root-mean-square differences of WV and PD are 1.86 kg m−2 and 11.4 mm, respectively. Finally, the retrieved results from the CMR brightness temperature along-track data using the retrieval models and the AMR along-track data are gridded by the inverse distance weighted method. Their monthly-mean spatial distributions are compared in order to investigate the applicability of the retrieval models globally, namely, beyond locations of the data pairs in the validation dataset. The comparison shows that the retrieval models are applicable in most open ocean areas, and that the latitude and temporal inhomogeneity of the crossover point distribution in the fitting dataset does not lead to obvious latitude and temporal inhomogeneity in the gridded data.
