Search

Satellite infrared detectors cannot penetrate clouds, especially precipitating clouds. Improving precipitation estimation accuracy based on infrared brightness temperature has always been important but challenging. In this paper, based on the infrared brightness temperature of the Advanced Geosynchronous Radiation Imager (AGRI) onboard China's Feng‐Yun 4A satellite, we develop and evaluate a new precipitation estimation method. First, using static data, physical characteristics of clouds, cloud image texture features, temporal motion features, and AGRI infrared channel brightness temperature, we construct features for a machine learning model. Then, we develop precipitation estimation methods. Precipitation is estimated in two steps: classification and regression. We employ a random forest classification model to identify whether there is precipitation in a given field of view. If there is precipitation, a multi‐model ensemble regression learning method is used to estimate the areas with this precipitation. The ensemble learning method uses convex optimization to integrate prediction results based on the optimization of hyperparameters of five basic models (i.e., those of random forest, XGBoost, LightGBM, decision tree, and extra tree models). Furthermore, two regression stacking ensemble models—the Least Absolute Shrinkage and Selection Operator (herein referred to as Stacking1‐LASSO) and K‐nearest neighbor (herein referred to as Stacking2‐KNN)—are used to predict the results of the aforementioned basic models. The results of basic models are used as inputs of these two stacking models. Finally, based on the Integrated Multi‐satellitE Retrievals for GPM (IMERG) precipitation product and rain gauge precipitation data, we conduct precipitation estimation experiments and evaluate our methods. The results show that ensemble learning models have greater accuracy in estimating precipitation than the basic models. When using IMERG precipitation as the target precipitation, ensemble learning models can estimate the central area of heavy precipitation during typhoons Ampil and Maria. The ensemble learning estimation effect is better than that of Stacking2‐KNN. Moreover, when rain gauge data is used as the target precipitation, ensemble learning can also estimate the center of heavy precipitation and with good consistency with recorded satellite brightness temperature data. Key Points: A convex optimization weighted and stacked ensemble learning method for satellite data estimation of precipitation is developedNear real‐time precipitation estimates can be obtained throughout the day using only geographic information and Advanced Geosynchronous Radiation Imager infrared brightness temperature dataThe proposed ensemble learning method can estimate the central areas of typhoons and short‐term heavy rainfall [ABSTRACT FROM AUTHOR]

Historically, the long wavelength side (6–7.5 μm) of the mid‐infrared water vapor absorption band has been used for imaging from the geostationary perspective. This began with the 6.4 μm band on Europe's Meteosat‐1. While geostationary sounders for moisture profiling, including China's hyperspectral resolution infrared sounder and a planned sounder from Europe, are or will be measuring the short wavelength side of the water vapor band, this is not the case for geostationary imagers. Shorter wavelength (5–6 μm) spectral bands for imaging applications should be considered for observing moisture in the mid and lower troposphere because of several potential advantages offered by this spectral range. The short wavelength side of the water vapor band contains fewer additional absorbing gases that overlap the water vapor absorption lines. In addition, the shorter wavelengths would show less diffraction blurring which could enable finer spatial resolution for turbulence detection. This study considers some of the differences and potential advantages of the spectral information in the short wavelength side of the water vapor absorption band from the perspective of geostationary imaging. For dry conditions land heating could impact the qualitative use of observed brightness temperatures at 5.1 μm, while the solar reflection component over most clear‐sky scenes is small for a 5.1 μm band and essentially zero for a 5.6 μm band. Plain Language Summary: Long before images are sent back to Earth from satellites, the needed spectral bands have to be specified. The band specification includes attributes such as the wavelength centers, widths, and signal‐to‐noise ratio. This paper investigates the rationale for including data from a region of the electromagnetic spectrum that has not been used for imaging from geostationary orbit. If implemented, these short wavelength water vapor bands would provide unique information regarding low‐level moisture and mid‐level turbulence. This paper quantifies the expected reflected component from the sun and the influence of land heating on observations, both of which are relatively larger at shorter wavelengths. Key Points: Historically, the long wavelength side of the mid‐infrared water vapor absorption band has been used for imagingShort wavelength (near 5 μm) spectral bands should be considered, especially with finer spatial resolutionsOver most clear‐sky surfaces, the reflective component ranges from small to zero for 5.1 and 5.6 μm bands [ABSTRACT FROM AUTHOR]