11:15 〜 11:30
[AAS07-09] JAXA EarthCARE MSI L2雲プロダクトの早期結果及び検証
キーワード：衛星観測、はくりゅう、全天カメラ画像、雲光学的厚さ、雲量
Our cloud phase & cloud properties algorithm for Earth Clouds, Aerosols and Radiation Explorer/multi-spectral imager (EarthCARE/MSI) L2 cloud products has been updated to v1.0 by early 2025. During the updates of the algorithm from v0.4 to v1.0, we fixed several bugs and applying new LUTs for both water and ice cloud retrievals.
After the launch of EarthCARE which scheduled in May of 2024, we started the validation of MSI L2 cloud products based on the early phase version (vAa-vAf) of MSI L1 cloud data. Firstly, the cloud phase algorithm (CLAUDIA) cloud cover validation used three ground-based sky camera system at Shibuya (Tokyo), Kumamoto (Kyushu), and Iriomote island (Okinawa). We developed a machine learning based algorithm of the sky camera (2-pi camera) images to discriminate cloud cover with clear area. Logistic regression method and XGBoost were used in our algorithm, and our results from the algorithm have shown very high accuracy on images of both day and night. Besides, our results of confusion matrix, Receiver Operating Characteristic (ROC) and Area Under the Curve (AUC) also have shown high precision of the cloud cover discrimination algorithm. For better cloud monitoring in the urban area of Tokyo, we added a new whole-sky camera system (Prede PSV-50) in early September 2024, which has a foldable shadow blade inside of the dome to shade the sun. During nighttime, the shadow blade is folded automatically. In the future, we are planning to add this new camera system to Iriomote as well.
In this presentation, we are going to show some early phase results of the JAXA EarthCARE MSI L2 Cloud Products, including both warm water clouds and ice clouds. Meanwhile, the detailed results of MSI L2 cloud cover validation, as well as the initial results of MSI L2 cloud optical thickness validation with MODIS-Aqua cloud products MYD06, will be shown in the presentation.
After the launch of EarthCARE which scheduled in May of 2024, we started the validation of MSI L2 cloud products based on the early phase version (vAa-vAf) of MSI L1 cloud data. Firstly, the cloud phase algorithm (CLAUDIA) cloud cover validation used three ground-based sky camera system at Shibuya (Tokyo), Kumamoto (Kyushu), and Iriomote island (Okinawa). We developed a machine learning based algorithm of the sky camera (2-pi camera) images to discriminate cloud cover with clear area. Logistic regression method and XGBoost were used in our algorithm, and our results from the algorithm have shown very high accuracy on images of both day and night. Besides, our results of confusion matrix, Receiver Operating Characteristic (ROC) and Area Under the Curve (AUC) also have shown high precision of the cloud cover discrimination algorithm. For better cloud monitoring in the urban area of Tokyo, we added a new whole-sky camera system (Prede PSV-50) in early September 2024, which has a foldable shadow blade inside of the dome to shade the sun. During nighttime, the shadow blade is folded automatically. In the future, we are planning to add this new camera system to Iriomote as well.
In this presentation, we are going to show some early phase results of the JAXA EarthCARE MSI L2 Cloud Products, including both warm water clouds and ice clouds. Meanwhile, the detailed results of MSI L2 cloud cover validation, as well as the initial results of MSI L2 cloud optical thickness validation with MODIS-Aqua cloud products MYD06, will be shown in the presentation.