Scheme for detection of low clouds from geostationary weather satellite imagery

• We develop a low cloud detection algorithm for MTSAT-2.
• We aim at constructing a simple and convenient scheme.
• Comparison to GPV data enables extensive data collection for threshold determination.
• Verification with radiosonde data proves the adequacy of this scheme.

A simple and practical scheme for low cloud detection at sea based on geostationary weather satellite data is proposed. The scheme consists of several threshold discrimination tests, and time-consuming procedures are eschewed in order to enable near-real-time analysis. This scheme also minimizes the use of data other than from geostationary weather satellites, leading to a convenient low cloud detection procedure without ancillary data. Careful investigation of the radiative properties of low water clouds from radiative transfer simulation and satellite observations enables full utilization of the characteristics of satellite data and realization of the simple scheme. The threshold values, which should possess high generality, for the discrimination tests are obtained from statistical comparisons of Multi-function Transport Satellite-2 data to grid point value data, which allows for extensive data collection and eliminates the localities and anomalies. Verification by comparisons with radiosonde and lidar on satellite suggests that results obtained from the proposed low cloud detection scheme are reasonable.