VALIDATION OF DIFFERENT CLOUD PHASE RETRIEVAL METHODS FROM SEVIRI USING GROUND-BASED CLOUD RADAR AND LIDAR OBSERVATIONS

Four satellite cloud phase determination methods from SEVIRI were evaluated by comparing one year of retrievals (May 2004-April 2005) with ground-based observed cloud phase from cloud radar and lidar at Cabauw, the Netherlands. The aim of this research was to assess which cloud phase determination method was suited best to derive cloud phase climatology. Differences between these methods were quantified by studying seasonally averaged liquid water phase ratio; the skill of the methods was assessed using weighted correlation coefficients between daily liquid water phase ratio from SEVIRI and ground-based observations. For the temperature thresholding methods, using the commonly used threshold of 260 K, the difference in liquid water phase ratio is 30-35 % during the summer and autumn months, decreasing to ~20 % during the remainder of the period investigated. Further, brightness temperature difference 10.8-12.0 μm (BTD10.8-12.0) has a poor quality in deriving cloud phase climatology. Thirdly, the difference in seasonal liquid water phase ratio between CTT and BT10.8 is only a few percent, which raises the suggestion that correction for cloud emissivity does not play a large role at the threshold of 260 K. Finally, 0.6 and 1.6 μm reflectance ratio shows an increasing difference in seasonal liquid water phase ratio with ground-based observations towards the winter season due to unfavorable viewing geometries