Daytime Cloud Mask for Himawari-8/9 (HIM-CLD-Day_Mask)

Brief description

This cloud mask collection was generated for daytime Himawari-8/9 imagery using a newly developed algorithm by CSIRO (Commonwealth Scientific and Industrial Research Organisation), collaboratively among research teams, for the Australian continent and surrounding waters. The product was extensively validated against space-borne LiDAR to ensure its quality. The product is provided here in a regular latitude/longitude grid. It is also available in the original Himawari (geostationary) projection (WGS84, sub-satellite longitude = 140.7°E and satellite altitude = 35785863 meters). Please contact the authors for accessing the data, which is located on the Australian National Computational Infrastructure (NCI). Currently, a near-real-time algorithm/processor is being developed to improve data availability, and to extend production to Himawari-10 (2028-2045).

Notes

Binary mask, 0 = cloud absent (clear) and 1 = cloud present. The binary mask is 1 if there is any non-zero amount of cloud in the pixel. A separate confidence variable is provided to specify how confidently the cloud mask is determined. This product is for day time only (when solar zenith angle is less than 75 degrees). (currently defined so that accuracy at cloudy and clear condition is balanced. examples using confidence).

Lineage

This cloud mask collection was generated using an algorithm independently developed by CSIRO based on a time-series analysis approach. A previous version applied to Advanced Along-Track Scanning Radiometer (AATSR) imagery (European Space Agency, Dual-View polar orbiting scanner) showed good performance (Qin et al., 2015). On this basis, the algorithm was improved and adapted to the Himawari series. The cloud mask data was validated extensively against the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) instrument aboard the CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation) satellite over nearly 6 years with over 200 million match-up samples. The validation demonstrated that the product provided cloud masking at an overall accuracy of 98% for all clouds; See Qin et al. (2019) for full details.

Notes

We at the Terrestrial Ecosystem Research Network (TERN) acknowledge the Traditional Owners and Custodians throughout Australia, New Zealand and all nations. We honour their profound connections to land, water, biodiversity and culture and pay our respects to their Elders past, present and emerging.

This work was funded by TERN, an Australian Government NCRIS (National Collaborative Research Infrastructure Strategy) enabled project, and is supported using TERN infrastructure. The research was undertaken at CSIRO (Commonwealth Scientific and Industrial Research Organisation) collaboratively among research units (Environment, Energy, Agriculture and Food). The authors acknowledge the resources and services received from NCI and CSIRO HPC (High Performance Computing). We appreciate the support by Japan Meteorology Agency (JMA) and Australian Bureau of Meteorology (BoM) in providing Himawari data.

The Himawari satellite series offers a unique opportunity to monitor sub-daily processes on Earth and in its atmosphere over Asia and Oceania, due to its unprecedented 10-minute temporal resolution as well as improved spatial and spectral resolutions compared to the previous generations of Japanese geostationary satellites and sensors (i.e., MTSAT series), which Australian researchers are kindly provided access through a JMA-BoM agreement.

However, almost all applications of optical remote sensing data, such as that of the Himawari series, require reliable cloud masking. This product provides a high-quality cloud mask for daytime Himawari imagery, enabling improved applications of daytime Himawari data. Due to its high temporal frequency, this product can also be used to cloud-mask other optical remote sensing imagery of compatible spatial resolution (i.e., >= 500 m pixel resolutions).