A new method for assessing and communicating aviation in-flight icing hazards has been proposed. This methodology creates a simple numerical index for quantifying hazard severity. The index is traceable to flight-level meteorology and aircraft-specific, icing-induced reductions in aircraft performance. It also provides a connection to a statistical data base of icing meteorology. This system will clarify the terminology used to describe the degree of danger posed by specific meteorological conditions. The relationship between hazard severity and meteorology is related by measured ice accumulation rates observed on a standard airfoil under prescribed conditions. This system has greater fidelity than the existing system and is applicable to all types of air vehicles.

Figure 1. A Meteorological Matrix would be used to characterize the range of meteorological conditions that pose in-flight icing hazards. The shaded region is a fictitious example of a boundary defining the region for safe operation of a specific aircraft.
The proposed numerical index is based on a multidimensional matrix representation of meteorological parameters that pertain to icing. For example (see Figure 1), suppose that ice-accretion rates for a given aircraft could be determined from three parameters; the outside air temperature, the fraction of cloud water droplets with diameters >100 μm, and the liquid density (the volume of water per unit volume of air). The three-dimensional space for these three parameters would be segmented into cells, each representing a unique meteorological state. Each cell could be assigned a probability of occurrence estimated from meteorological data bases. Aircraft manufacturers would then be able to specify surfaces in the three-dimensional parameter space that bounds safe operating conditions for each of their aircraft for various ranges of exposure times. Thus, the meteorological matrix concept would provide traceability among meteorological conditions, aircraft performance, and cumulative probabilities of occurrence of icing.

To reference each level of the proposed index to the degree of hazard, the index would be related to measured rates of ice accumulation on a standard wing cross section. The rates would be measured over a wide range of meteorological conditions for a standard set of flight conditions (such as airspeed and angle of attack). Aerodynamic modeling software could then be used to translate the observed icing phenomena to commercial airfoil shapes with some confidence.

Figure 2. Hazard Indices ranging from 3 to 9 have been placed in the cells in one plane of a meteorological matrix. The hazard indices shown here constitute a fictitious example.
The proposed index would feature some number of levels — possibly 12 — chosen to increase the fidelity of reporting beyond that of the current four-level system, without making the levels so narrow that the differences between them could not be reasonably measured. The levels would be assigned to cells in meteorological matrices (see Figure 2). The twelve-level scale would be related to the present four-level system in the following way: Zero would represent meteorological conditions that do not induce icing; three through six would correspond to the "light" level; seven through nine would correspond to the present "moderate" level; and 10 to 12 would correspond to the present "severe" level. The correlations between measured icing rates, the present four levels, and the proposed index would be established in a consensus process that would involve airlines, pilot organizations, government, and aircraft manufacturers.

This work was done by Steven J. Walter of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp under the Physical Sciences category. NPO-20465