Global Classification Dataset of Daytime and Nighttime Marine Low-cloud Mesoscale Morphology

<p>The global classification dataset of daytime and nighttime marine low-cloud mesoscale morphology with six cloud types (Solid stratus, Closed MCC, Open MCC, Disorganized MCC, Clustered Cu and Suppressed Cu). The spatial resolution is 1^o &times; 1^o&nbsp;and the temporal resolution is 5 minutes for the years 2018-2022. They were established based on a deep learning model ResNet-50. Trained on daytime radiance data from MODIS (Moderate Resolution Imaging Spectroradiometer) and daytime retrieved COT (Cloud Optical Thickness), this model achieved a high prediction accuracy and can be applied to nighttime cloud classification. For a detailed introduction to the model, please refer to our article.</p> <p>&nbsp;</p> <h2>Technical info</h2> <p><strong>Product information</strong></p> <ul> <li><strong>File &lsquo;day_xxxx_all.h5&rsquo;:</strong> Daytime classification of global marine low-cloud morphology for the year xxxx, with a spatial resolution of 1&deg;&times;1&deg; and a temporal resolution of 5 minutes&nbsp; <ul> <li>date: time of the 1&deg;&times;1&deg; box, format: 'YYYYDDD.HHHH'</li> <li>lon: central longitude (-180, 180)</li> <li>lat: central latitude (-60, 60)</li> <li>cat: category of the cloud morphology. The numbers 0-5 represent each of the six categories: 0-Solid stratus, 1-Closed MCC, 2-Open MCC, 3-Disorganized MCC, 4-Clustered Cu, 5-Suppressed Cu</li> <li>cert: model certainty, the probability that this cloud morphology belongs to the assigned category</li> <li>low_cf: the cloud fraction of low clouds</li> <li>COT_CNN: average cloud optical thickness (COT), retrieved using TIR-CNN model from Wang et al. (2022)</li> <li>CER_CNN: average cloud effective radius (CER), retrieved using TIR-CNN model from Wang et al. (2022), in the unit of &mu;m</li> <li>LWP_CNN: average cloud liquid path (LWP), calculated from COT_CNN and CER_CNN, in the unit of g/㎡</li> <li>Sensor_zenith: scene average sensor zenith angle, from MODIS MYD021, in the unit of degree (&deg;)</li> </ul> </li> </ul> <div> <ul> <li><strong>File 'night_xxxx_all.h5':</strong> Nighttime classification of global marine low-cloud morphology for the year xxxx, with a spatial resolution of 1&deg;&times;1&deg; and a temporal resolution of 5 minutes&nbsp;&nbsp; <ul> <li>same variables as daytime</li> </ul> </li> </ul> <ul> <li><strong>File 'example.xlsx':</strong> A sample of the variable data from our cloud classification dataset, showcasing the classification results of a MODIS granule captured on January 1, 2018, at 00:25 UTC. This sample is provided to help users better understand the content of our dataset.</li> </ul> <p>&nbsp;</p> <p><strong>Training, Validation, Test dataset</strong></p> <ul> <li>Originating from the same classification dataset, same variables, only differ in sample size</li> <li><strong>Files 'training_dataset.h5', 'validation_dataset.h5', and 'test_dataset.h5'</strong><strong>:</strong> <ul> <li>date: time of the 128&times;128 pixels, format: 'YYYYDDD.HHHH'</li> <li>lat: central latitude of the 128&times;128 scene</li> <li>lon: central longitude of the 128&times;128 scene</li> <li>cat: category of the cloud morphology. The numbers 0-5 represent each of the six categories: 0-Solid stratus, 1-Closed MCC, 2-Open MCC, 3-Disorganized MCC, 4-Clustered Cu, 5-Suppressed Cu</li> <li>CTH: cloud top height, in-cloud average value, in the unit of km</li> <li>COT_retrieved: cloud optical thickness (COT), retrieved using TIR-CNN model from Wang et al. (2022), 128&times;128 pixels</li> <li>LWP: cloud liquid path (LWP) from MODIS MYD06, in-cloud average value, in the unit of g/㎡</li> <li>Sensor_zenith: scene average sensor zenith angle, from MODIS MYD021, in the unit of degree (&deg;)</li> <li>emis_29: radiance data from thermal infrared channel 29 (8.7&mu;m), 128&times;128 pixels</li> <li>emis_31: radiance data from thermal infrared channel 31 (10.8 &mu;m), 128&times;128 pixels</li> <li>emis_32: radiance data from thermal infrared channel 32 (12.0 &mu;m), 128&times;128 pixels</li> <li>i: the row number of the top-left pixel of 128 &times;128 scene in the MODIS granule</li> <li>j: the column number of the top-left pixel of 128 &times;128 scene in the MODIS granule</li> </ul> </li> </ul> </div>