Assessment of Aircraft Icing Potential Using Satellite Data

Abstract

This paper explores the potential of using satellite data to develop a climatology of aircraft icing probability in oceanic regions. The datasets employed in this analysis are: the Nimbus-7 Scanning Multichannel Microwave Radiometer (SMMR) microwave radiances; the U.S. Air Force Three-Dimensional Nephanalysis (3DNEPH); the European Centre for Medium-Range Weather Forecasts (ECMWF) initialized analyses; and the High-Resolution Infrared Sounder 2-Microwave Sounding Unit (HIRS2-MSU) satellite radiances. This analysis focuses on the midlatitude regions of the North Atlantic Ocean, encompassing the paths of most transatlantic flights between the United States and Europe. Estimates of cloud temperature, cloud horizontal extent, liquid water path, cloud depth, cloud liquid water content, drop-size characteristics, and precipitation characteristics are obtained from these datasets. A total of 14% of low-level clouds are found to have average temperatures between ?2° and ?36°C, while 99% of midlevel clouds fall in this temperature range. More than half of the cloud decks have a horizontal areal extent of less than (46 km)2, and no cloud decks exceed (442 km)2 in horizontal extent. The mean low-cloud depth is determined to be about 1000 m and the midcloud depth to be about 1950 m. A maximum value of 800 g m?2 supercooled liquid water path is obtained considering single-layer cloud decks with 100% areal coverage. The average values of the liquid water path for the supercooled middle and low clouds are 62 and 92 g m?2, respectively. Using an assumed shape for the vertical profile of liquid water content, average values of 0.095 g m?3 for low-level clouds and 0.043 g m?3 for midlevel clouds are determined. The maximum value of liquid water content is determined to be 1 g m?3. Assessment of aircraft icing potential for these clouds is made by first determining the probability of icing and then determining the ice accretion per icing encounter. The joint probability of an icing encounter with a specified mass of ice accretion can then be used to quantify the icing threat. This analysis shows maximum icing probability of 4.3% occurring at altitudes between 1.3 and 2.5 km, with icing probabilities less than 1% at altitudes below 200 m and above 5 km. Assuming a collection efficiency of unity, the average ice accretion per icing encounter is determined to be about 5 kg m?2, with a maximum average value of 13 kg m?2. A maximum value of ice accretion as large as 250 kg m?2 is determined near the 1400-m altitude. Under average conditions, the maximum icing threat is seen to occur at an altitude of 1700 m, corresponding to a 4.3% probability of an ice accumulation of 13.5 kg m?2. A discussion of errors and uncertainties associated with this methodology is given.