Search

This study introduces a data-to-data (D2D) translation approach that utilizes a conditional adversarial learning framework to generate hypothetical data at the meteorological imager (MI) sensor on the communication, ocean, and meteorological satellite (COMS). The proposed D2D model produces virtual 10-min data from actual 30-min data by exploiting the 10-min temporal resolution (TR) of the advanced MI (AMI) on GEO-KOMPSAT-2A during the overlapping observation period of the two satellites. Specifically, the D2D model uses one visible (VIS) at 0.64 μm channel and four infrared channels (3.8, 6.9, 10.8, and 12.3 μm) from AMI of the actual 30-min data from April 2020 to April 2022 to train and test the model. Subsequently, the D2D model is applied to simulate hypothetical 10-min-TR COMS data using the 30-min COMS observation data from September 2019 to March 2020 during the coexistence period of the two satellites. Regression-calibrated COMS data alleviated the spectral response function differences between MI and AMI sensors. The proposed D2D method exhibits excellent statistical performance, with an average root-mean-square error of 0.056 for the VIS channel and 3.237 K, 1.005 K, 3.251 K, and 3.184 K for 3.8 μm, 6.9 μm, 10.8 μm, and 12.2 μm channels, respectively. The findings of this study are expected to facilitate various types of remote sensing research and applications using long-term data with AMI and AMI-like past MI data.

True-color imagery is essential for an intuitive comprehension of atmospheric data. However, the Geostationary Environment Monitoring Spectrometer (GEMS) of the geostationary Korea multipurpose satellite (GK) 2B lacks green and red bands, which limits its ability to monitor atmospheric environments. To mitigate this issue, we suggest an innovative method of generating virtual GEMS green and red bands using conditional generative adversarial networks with data observed in the blue-green-red (RGB) bands of the Advanced Meteorological Imager sensor, a payload of the GK-2A satellite. The paired datasets of the AMI blue band and the AMI RGB bands were used to train and test the data-to-data (D2D) translation model. Using the GEMS blue band as input data, the D2D model generated GEMS hypothetical radiance data at the green and red bands. Our results show that the D2D model generated hypothetical GEMS green and red bands with outstanding performance. The averaged values of the correlation coefficient, root-mean-square error, and bias between the observed and D2D-generated GEMS blue band were 0.999, 3.450 W/cm2/cm/sr, and −1.858 W/cm2/cm/sr, respectively. This research is expected to significantly contribute to the monitoring and comprehension of atmospheric environments in Asia and potentially improve the GEMS's global ability to monitor air quality. Additionally, the proposed method has the potential to enhance the capabilities of other satellites with limited spectral bands.

Cirrus cloud contributes significantly to earth's radiation budget and the greenhouse effect. The Advanced Himawari Imager (AHI) onboard the Himawari-8 satellite lacks a 1.37 μm band, sensitive to monitoring cirrus clouds. This article proposed a conditional generative adversarial network-based data-to-data translation (D2D) model to generate a hypothetical AHI 1.37 μm band. For training and testing the D2D model, the Geo-Kompsat-2A Advanced Meteorological Imager (AMI) 1.37 μm bands and other highly correlated bands to cirrus from July 24, 2019 to July 31, 2020, were used. The D2D model exhibited a high level of agreement (mean of statistics: correlation coefficient (CC) = 0.9827, bias = 0.0004, and root-mean-square error (RMSE) = 0.0086 in albedo units) between the observed and D2D-generated AMI 1.37 μm bands from validation datasets. The application of the D2D model to the AHI sensor showed that the D2D-generated AHI 1.37 μm band was qualitatively analogous to the observed AMI 1.37 μm band (average of statistics: bias = 0.0026, RMSE = 0.0191 in albedo units, and CC = 0.9158) on the 1st, 15th, and 28th of each month of 2020 in the common observing regions between Korea and Japan. The validation results with the CALIPSO data also showed that the D2D-generated AHI 1.37 μm band performed similarly to the observed AMI 1.37 µm band. Consequently, this article can significantly contribute to cirrus detection and its application to climatology.

True-color imagery of satellites provides various atmospheric and surface information for intuitive understanding and visualization. This article presents a conditional generative adversarial network method for generating daytime and nighttime hypothetical visible (VIS) bands of the advanced meteorological imager (AMI) sensor onboard the geostationary Korea multipurpose satellite. The AMI datasets in the form of albedo and brightness temperature (${{\boldsymbol{T}}}_{\boldsymbol{B}}$) were normalized and denormalized between 0 and 1 data-to-data (D2D) translation. The D2D model was trained and tested using data pair of the albedos at AMI VIS bands and AMI infrared (IR) bands or ${{\boldsymbol{T}}}_{\boldsymbol{B}}$ differences between two AMI IR bands. The constructed D2D model showed that the statistical results of bias, root-mean-square-error, and correlation coefficient between the observed and D2D-generated AMI VIS bands during daytime were −0.006 and 0.047 in albedo, and 0.941 for the blue band; −0.007 and 0.05 in albedo and 0.939 for the green band; −0.01 and 0.061 in albedo, and 0.917 for the red band, respectively. The proposed D2D method is being officially used by the Korea meteorological administration. Except for simulating desert areas as clouds at night, the D2D model demonstrated excellent performance, generating hypothetical AMI VIS bands at day and night. Consequently, this article could significantly contribute to the monitoring and understanding of meteorological phenomena over one-third of the Earth. Additionally, the method can be extended to other geostationary weather satellites, including Himawari-8, Fengyun-4A, meteosat third generation, and geostationary operational environmental satellites.

This study presents a deep learning model, using the conditional generative adversarial nets (CGAN) technique, that can produce daytime visible (VIS) band information, mimicking a narrow band sensor, by combining VIS and infrared (IR) broadband measurements by different sensors. The real-observed datasets of the Geostationary Ocean Color Imager (GOCI) and meteoritical imager (MI) sensors onboard the Communication, Ocean, and Meteorological Satellite were used for training and testing our CGAN model over the Yellow Sea and Bohai Sea. The trained and tested CGAN model was then applied to generate daytime GOCI VIS and near IR (NIR) bands (0.412 to 0.865 μm) using daytime MI VIS (0.675 μm), shortwave IR (3.75 μm), and longwave IR bands (10.8 and 12.0 μm) and the differences between them as input data. GOCI and MI data were collected from January 2017 to December 2018 using 705 images of 256 × 256 pixels for the training and 44 images for the model test. The results are statistically favorable (i.e., bias = −0.013 (in a reflectance unit from 0 to 1), root-mean-square error = 0.112, mean absolute error = 0.076, agreement index = 0.945, and correlation coefficient (CC) = 0.809 for daytime reflectance in the GOCI VIS 0.49-μm band) between the real GOCI VIS band observation and the CGAN-generated simulation. Our CGAN-based model showed high CC and favorable results in the GOCI VIS and NIR bands. Consequently, our study demonstrates the possibility of applying a deep learning technique to improve the temporal resolution for ocean color studies using the GOCI sensor.

The visible band of satellite sensors is of limited use during the night due to a lack of solar reflection. This study presents an improved conditional generative adversarial networks (CGANs) model to generate virtual nighttime visible imagery using infrared (IR) multiband satellite observations and the brightness temperature difference between the two IR bands in the communication, ocean, and meteorological satellite. For the summer daytime case study with visible band imagery, our multiband CGAN model showed better statistical results [correlation coefficient (CC) = 0.952, bias = -1.752 (in a digital number (DN) unit from 0 to 255, converted from reflectance from 0 to 1), and root-mean-square-error (RMSE) = 26.851 DN] than the single-band CGAN model using a pair of visible and IR bands (CC = 0.916, bias = -4.073 DN, and RMSE = 35.349 DN). The proposed multiband CGAN model performed better than the single-band CGAN model, particularly, in convective clouds and typhoons, because of the sounding effects from the water vapor band. In addition, our multiband CGAN model provided detailed patterns for clouds and typhoons at twilight. Therefore, our results could be used for visible-based nighttime weather analysis of convective clouds and typhoons, using data from next-generation geostationary meteorological satellites.