A newly designed thermal/vacuum (T/V) prelaunch multiangle calibration test is proposed for Microwave Humidity and Temperature Sounder (MWHTS) onboard the Feng Yun 3D satellite. Compared with other existing prelaunch calibration tests designed at fix scan angle, this test is first proposed to analyze the measurements of MWHTS from multiscan angles. A set of comprehensive calibration method is established and used for calibration data processing. The high-quality data preprocessing methods are performed. Essential calibration coefficients such as cold bias, hot bias, and nonlinearity coefficients, and calibration results such as calibration residual error, sensitivity, and calibration accuracy of MWHTS are all obtained and analyzed in MWHTS's six observing positions. The measured sensitivity value of channel 2 is 1.5 K while the measured sensitivities of the others range from 0.18 to 0.57 K. The calibration accuracy of channel 2 is 1.5 K while the calibration accuracies of the others range from 0.32 to 0.7 K. According to the discrepancy of multiangle calibration test results, the radiation leakage and the response lag for special channels are discovered for the first time in T/V test, and the reasons of these two special phenomena are fully analyzed. Moreover, the measurements of the T/V test with fast changing instrument temperatures are analyzed and the applicability of nonlinearity coefficients obtained in T/V test for in-orbit calibration is verified. In conclusion, MWHTS meets the basic design requirements of sensitivity and calibration accuracy, the changing of the MWHTS observing positions does not impact the calibration coefficients and results, and the nonlinearity coefficients obtained in T/V test are valid when instrument temperature experiences some fast changing.

On November 15, 2017, the fourth satellite, Feng Yun-3D (FY-3D), of the new generation of Chinese polar orbit meteorological satellites was successfully launched into orbit. As one of the key payloads, the second-generation Microwave Humidity Sounder (MWHS II) has 15 channels for observing atmospheric temperature and moisture. In the ground data processing system, the calibration is completed for every scanning period, and then, brightness temperature (TB) data are obtained with calibration coefficients. In order to analyze the performance of MWHS II, one year of in-orbit data are used to monitor the trend of internal blackbody target temperature, instrument temperature, calibration counts, and automatic gain control (AGC), and the results indicate whether the instrument is working well. The instrument noise equivalent differential temperature (NEDT) is calculated, and the value satisfies the design requirements. Two methods are used to validate FY-3D MWHS II measurements. One is the simultaneous nadir overpass (SNO) approach: in this method, ATMS is chosen as the reference sensor, and the standard deviations for five humidity sounding channels are all less than 1.3 K. In the alternate approach, a comparison of the observed and simulated TB of MWHS II is carried out between FY-3C and FY-3D. The results also demonstrate a consistent performance between the two instruments.

The microwave humidity and temperature sounder (MWHTS) on the Fengyun (FY)-3C satellite measures the outgoing radiance from the Earth’s surface and atmospheric constituents. MWHTS, which makes measurements in the isolated oxygen absorption line near 118 GHz and the vicinity of the strong water vapor absorption line around 183 GHz, can provide fine vertical distribution structures of both atmospheric humidity and temperature. However, in order to obtain the accurate soundings of humidity and temperature by physical retrieval methods, the bias between the observed and simulated radiance calculated by the radiative transfer model from the background or first guess profiles must be corrected. In this study, two bias correction methods are developed through the correlation analysis between MWHTS measurements and air mass identified by the first guess profiles of the physical inversion; one is the linear regression correction (LRC), and the other is the neural network correction (NNC), representing the linear and nonlinear relationships between MWHTS measurements and air mass, respectively. The correction methods have been applied to MWHTS observed brightness temperatures over the geographic area (180° W–180° E, 60° S–60° N). The corrected results are evaluated by the probability density function of the differences between corrected observations and simulated values and the root mean square errors (RMSE) with respect to simulated observations. The numerical results show that the NNC method has better performance, especially in MWHTS Channels 1 and 7–9, whose peak weight function heights are close to the surface. In order to assess the effects of bias correction methods proposed in this study on the retrieval accuracy, a one-dimensional variational system was built and applied to the MWHTS brightness temperatures to estimate the atmospheric temperature and humidity profiles. The retrieval results also show that NNC has better performance. An indication of the stability and robustness of the NNC method is given, which suggests that the NNC method has promising application perspectives in the physical retrieval.

In this article, two multitask deep learning models, multilayer perceptron (MLP) and convolutional neural networks (CNNs) are constructed to detect precipitation flags and retrieve precipitation rates simultaneously over the Northwest Pacific area. The retrieval results are verified by integrated multisatellite retrievals for GPM. Compared with using a single payload, the results show the retrieval advantages of incorporating two microwave spaceborne payloads (microwave humidity and temperature sounder and microwave radiation imager). In addition, each separated task in multitasking deep learning models can outperform the single-task models. The MAE values of MLP and CNN are decreased by 0.05 and 0.04 mm/h, respectively. Nevertheless, it is still a challenge to extend the application of these models, which demands extracting the features learned from a certain area to other areas characterized by different precipitation features. A flexible network framework is developed with two transfer learning (TL) methods (freeze and fine-tuning) to tackle this problem. The pretrained models (MLP and CNN) are trained from the Northwest Pacific area, whereas the transferred models are applied to the Northeast Pacific area. The encouraging results show that the performance of the TL model using the fine-tune method has exceeded that of the comparison model without TL, which is trained by four times more data. The main advantage of using TL is the time reduction and efficiency improvements.

As an important component of the Earth system, precipitation plays a vital role in regional and global water cycles. Based on Microwave Humidity and Temperature Sounder (MWHTS) onboard FY-3D satellite, four machine learning models, random forest regression (RFR), support vector machine (SVM), multilayer perceptron (MLP), and gradient boosting regression tree (GBRT), are implemented to retrieve precipitation rate, and verified with Integrated Multi-satellite Retrievals for GPM (IMERG). This paper determines the optimal hyperparameters of the machine models and proposes three linear combinations of MWHTS channels (183.31 ± 1.0–183.31 ± 3.0 GHz, 183.31 ± 1.0–183.31 ± 7.0 GHz, and 183.31 ± 3.0–183.31 ± 7.0 GHz), which can better characterize precipitation of different intensities. With the inclusion of three linear combinations, the performances of all four machine learning models are significantly improved. It is concluded that the RFR and GBRT have the best retrieval accuracy. Over ocean, the MSE, MAE, and R${}^2$ values of precipitation estimates using RFR are 1.75 mm/h, 0.44 mm/h, and 0.80, respectively, and are 1.80 mm/h, 0.45 mm/h, and 0.78 for GBRT. Simultaneously, this paper analyzes the retrieval results from the perspective of the different rain rates and temporal matching difference between MWHTS and IMERG data. The RFR and GBRT also maintain the best retrieval accuracy under the condition of Gaussian noise, indicating the relatively strong robustness and antinoise performance of ensemble learning models for precipitation retrieval.

The application of ground-based microwave radiometers (GMWRs), which provide high-quality and continuous vertical atmospheric observations, has traditionally focused on the indirect assimilation of retrieved profiles. This study advanced this application by developing a direct assimilation capability for GMWR radiance observations within the Weather Research and Forecasting Data Assimilation (WRFDA) system, along with a bias correction scheme based on the random forest technique. The proposed bias correction scheme effectively reduced the observation-minus-background (O−B) biases and standard deviations by 0.83 K (97.1 %) and 1.63 K (64.6 %), respectively. A series of 10 d experiments demonstrated that assimilating GMWR radiances improves both the initial conditions and the forecasts, with additional benefits from higher assimilation frequencies. In the initial conditions, hourly assimilation significantly enhanced low-level temperature and humidity fields, reducing the root-mean-square error (RMSE) for temperature by 6.32 % below 1 km and for water vapor mixing ratio by 1.98 % below 5 km. These improvements extended to forecasts, where 2 m temperature and humidity showed sustained benefits for over 12 h, and precipitation forecasts exhibited improvements to a certain extent. The time-averaged Fractions Skill Score (FSS) for 3 h accumulated precipitation within the 24 h forecasts increased by 0.02–0.04 (3.9 %–10.2 %) for thresholds of 3–6 mm.