Sensitivity Testing of Microwave Temperature Sounder-II Onboard the Fengyun-3 Satellite to Sea Surface Barometric Pressure Based on Deep Neural Network

Sea surface barometric pressure contributes to calculating the surface transmissivity so that the observations of Microwave Temperature Sounder-II channels with non-zero surface transmissivity contain the sea surface barometric pressure information. Meanwhile, all channels of Microwave Temperature Sounder-II are sensitive to sea surface barometric pressure due to the correlation between channels. However, the traditional sensitivity test method based on the radiative transfer model cannot characterize the sensitivity of Microwave Temperature Sounder-II to sea surface barometric pressure due to the correlations between channels. In this study, the relationship between atmospheric parameters and Microwave Temperature Sounder-II observations is studied by a deep neural network, and the deep neural network-based model for Microwave Temperature Sounder-II simulations is established. Then, the deep neural network-based test method for the sensitivity of Microwave Temperature Sounder-II to sea surface barometric pressure is developed, and the sensitivity test experiments are carried out. The experimental results show that the sensitivity of all channels of Microwave Temperature Sounder-II to sea surface barometric pressure is captured by the deep neural network-based test method. In addition, the retrieval experiments of sea surface barometric pressure using Microwave Temperature Sounder-II observations are carried out, and the retrieval results further validate the feasibility of the deep neural network-based test method.