Previous work on the use of satellite, radar, surface, and model data to identify locations of aircraft icing has shown that each of these individual data sources can do so, but that they also have their individual limitations.
The Integrated Icing Diagnostic Algorithm (IIDA) combines information from all four of these data sources by mapping satellite, surface and radar observation to the Rapid Update Cycle (RUC) grid. The algorithm uses interest maps and `fuzzy logic' equations to maximize the strengths and minimize the weaknesses of each data type. Initially a yes/no cloud-precipitation field is used to limit the icing diagnosis to places where clouds and precipitation exist. Then the physically based interest maps are used to create a floating point value of liquid water and supercooled large drop (SLD) potential at each grid point.
PIREP based verification results for the continental United States proved the IIDA to be the most efficient icing algorithm yet tested by NCAR, and to be very competitive with human generated AIRMETS. In this paper the use of the satellite, surface, radar, and RUC model data will be discussed. The extraction of the yes/no cloud-precipitation field will be shown, along with the theory and techniques that lead to the resulting icing potential fields. Comparisons of IIDA generated fields and data from a research aircraft will be used to demonstrate the strengths and weaknesses of the algorithm. Summary verification results for the IIDA vs. PIREPS and IIDA vs. the NASA Twin Otter research aircraft will be also be shown
The 8th Conference on Aviation, Range, and Aerospace Meteorology