Cloud-base height derived from a ground-based infrared sensor and a comparison with a collocated cloud radar

An improved algorithm to calculate cloud-base height (CBH) from infrared temperature sensor (IRT) observations that accompany a microwave radiometer was described, the results of which were compared with the CBHs derived from ground-based millimeter-wavelength cloud radar reflectivity data. The results were superior to the original CBH product of IRT and closer to the cloud radar data, which could be used as a reference for comparative analysis and synergistic cloud measurements. Based on the data obtained by these two kinds of instruments for the same period (January-December 2016) from the Beijing Nanjiao Weather Observatory, the results showed that the consistency of cloud detection was good and that the consistency rate between the two datasets was 81.6%. The correlation coefficient between the two CBH datasets reached 0.62, based on 73 545 samples, and the average difference was 0.1 km. Higher correlations were obtained for thicker clouds with a larger echo intensity. A low-level thin cloud cannot be regarded as a blackbody because of its high transmittance, which results in higher CBHs derived from IRT data. Because of a smaller cloud radiation effect for high-level thin cloud above 8 km, the contribution of the atmospheric downward radiation below the cloud base to the IRT cannot be ignored, as it results in lower CBHs derived from IRT data. Owing to the seasonal variation of atmospheric downward radiation reaching the IRT, the difference between the two CBHs also has a seasonal variation. The IRT CBHs are generally higher (lower) than the cloud radar CBHs in winter (summer). © 2018 American Meteorological Society.