To put this in perspective, by definition, freezing drizzle is precipitation with a droplet diameter of 200–500 microns while freezing rain is precipitation with a droplet diameter of 500–1,000 microns. Pilots must not be misled into thinking that taking off into freezing drizzle or light freezing rain is safe simply because the ground holdover time of an antiicing fluid has not been exceeded.

Although the term "ground deicing" is often used to refer to both deicing and antiicing, understanding the difference between the two procedures is important. Deicing removes frozen precipitation that adheres to the aircraft. Antiicing is required if the freezing precipitation continues (or is forecast to occur) after deicing and before takeoff.

A number of different methods, including hot air blowers and infrared heat, can be used for deicing, but the most common method of deicing is to spray the aircraft with a heated mixture of glycol fluid and water.

The glycol fluid used for deicing is called Type I fluid and is dyed orange for visibility. This glycol absorbs moisture and has a low freezing point. The higher the percentage of glycol in the glycol/water mixture used, the lower the outside temperature can be and still have the mixture be effective for cleaning the aircraft—but only to a point. Past this point, the effectiveness of Type I fluid decreases, and as a result, pure Type I is rarely used. A 50/50 mix of glycol and water is common and is good down to around –20 degrees C, and a 60/40 mix is effective to around –31 degrees C.

Heated Type I fluid mixed with water is very effective as a deicer, but it has very little ability to keep frozen contamination from reaccumulating. In other words, Type I fluid is not effective as an antiicer.

When antiicing is required, ground personnel must go back over the aircraft with an improved fluid known as Type II or Type IV. Type II is slightly amber in color, and Type IV is dyed emerald green. Both of these fluids are also glycols, but with thickening agents added that enable the fluid to be deposited in a thick film that can absorb a large amount of moisture before beginning to freeze. These fluids are designed to remain on the aircraft surfaces during ground operations, but they will shear and flow off during takeoff.

Unlike Type I fluid, Type II and Type IV are most effective as antiicers when unheated and undiluted and applied as the second step in a "two step" deicing/antiicing process. Although Type II and Type IV may be diluted and heated so that they may be used for deicing and antiicing in a single step, the dilution inevitably reduces the duration of the antiice protection. Additionally, unless the airplane has been deiced with Type I fluid before the application of Type II or Type IV, fluid residue may collect in aerodynamically "quiet" areas. Under certain temperature and high humidity conditions, this residue may rehydrate and freeze, potentially blocking or impeding unpowered flight controls. Single-step deicing/antiicing with diluted and heated Type II or Type IV, although common in some European countries, is seldom done in North America.

Pilots should be particularly aware of the danger of relying on Type I fluid for antiicing. Testing of Type I fluids conducted during the past year has revealed that the holdover times for Type I are even less than was previously believed. Simply stated, holdover time is the estimated time the application of antiicing fluid will keep frost or ice from forming and snow or slush from accumulating on the protected surfaces of the aircraft. Holdover time starts at the beginning of the final application of deicing/antiicing fluid, and it ends when the fluid loses its effectiveness.

SAE AEROSPACE develops the holdover-time tables for Type I, Type II, and Type IV fluids. The FAA and Transport Canada approve the tables.

Type II and Type IV fluids have been tested regularly for a number of years, and the numbers in the Type II and Type IV tables have been modified accordingly.

Type I fluids, however, have not been routinely tested, and the numbers in the Type I holdover-time table have not changed for many years. This year, as a result of new testing of Type I fluids, SAE AEROSPACE developed a new Type I table with holdover times that are significantly reduced from those in the old table. For example, for snow at 0 degrees C to –10 degrees C, the old holdover range of 6–15 minutes was reduced to 3–6 minutes. For snow colder than –10 degrees C, the new numbers were reduced to 2–4 minutes.

A majority of the members of the Holdover Time subcommittee of SAE AEROSPACE, including ALPA, voted to approve the new Type I table.

But Transport Canada, the FAA, and those U.S. carriers that routinely use Type I for antiicing opposed the new reduced numbers for Type I fluid. They recognized that if the new numbers are used, Type I cannot realistically be used for antiicing because the holdover time could expire well before takeoff.

Accordingly, Transport Canada and the FAA decided not to publish the new SAE-approved holdover-time table for Type I fluid and instead are allowing carriers to use the old table again this year.

The effect of this decision is to suppress the new information on the reduced holdover times. Pilots of those carriers who have elected to provide "full disclosure" will find both tables in their Flight Operations Manuals. All pilots, however, should be aware of the new Type I table (with significantly reduced holdover times) even if they are provided with only the same table used last year. The 6–15-minute holdover-time range specified for snow in the old table may be overly optimistic and misleading.

How should the holdover-time tables be properly used? Although the numbers in the tables reflect actual testing of the fluids, many factors determine holdover time in the real world. Among these are precipitation intensity, airplane surface temperature, relative humidity, and wind velocity and direction, all of which can reduce holdover time below the lowest time in the holdover time range.

Generally, the lower number in the holdover time range is used to indicate the estimated time of protection during moderate precipitation and the upper limit to indicate the same during light precipitation. The effects of heavy precipitation have not yet been sufficiently tested to enable computation of holdover times, but the times will always be something less than the lower limit of the holdover time range.

Although relating the length of holdover time to the intensity of the precipitation is an accepted practice, snow can fool pilots. The major determinant of when an antiicing fluid will fail and no longer be effective is the amount of water content in the snow. Yet, in the United States and Canada, the intensity of snow (and of freezing drizzle) is defined not by water content, but rather by prevailing visibility.

Light snow, as the National Weather Service defines it, is reported when the visibility is greater than or equal to 5/8 mile (1.0 km), moderate snow intensity is reported when the visibility is less than 5/8 mile but greater than ¼ mile (0.5 km), and heavy snow is reported when the visibility is less than ¼ mile. Because liquid water content and visibility cannot be correlated, light snow that is wet could result in shorter actual holdover times than those for moderate snow that has less water content.

Pilots must be extremely cautious and not assume that because snow is being reported as "light," the appropriate holdover time is the high number at the upper end of the holdover time range. In fact, because no commonly used method equates snow intensity with water content, a pilot has no accurate means of determining appropriate holdover time in "light" snow.

Pilots should therefore assume that in light snow, holdover time may well be at the lower end of the holdover time range. If Type I fluid has been used for antiicing, holdover time may be as short as 6 minutes if the old Type I holdover-time table is used or as little as 2 minutes if the new Type I holdover-time table is used.

At night, the situation becomes even more critical because the visibility determined by an observer or an ASOS will be double the daytime visibility for the same liquid-equivalent snowfall rate. This results in heavy snowfall intensity being converted to moderate, and moderate snowfall intensity being converted to light.

Pilots of Canadian carriers that use the holdover times published by Transport Canada may be misled into thinking that conditions of light snow do not require antiicing because Transport Canada has chosen to put the label "Moderate Snow" on the snow column in the holdover-time tables. Transport Canada is alone in this practice. Holdover-time tables published in the United States and Europe correctly refer only to "Snow," which can be of any intensity from "light" to "moderate." If snow is falling, the aircraft should be antiiced before takeoff.

An understanding of how snow (and freezing drizzle) can fool pilots is essential if pilots are to properly use the holdover-time tables. Regulations in both Canada and the United States require that holdover times be used in conjunction with a visual check. The basic rule is that anytime holdover-time procedures are in effect, pilots must make a pretakeoff check of the aircraft’s wings and other critical surfaces.

In the United States, if the takeoff can be made within the holdover time, the flight crew may accomplish the pretakeoff check from the cockpit. If the holdover time will be exceeded, a pretakeoff contamination check must be made either from outside the aircraft or from the cabin within 5 minutes of takeoff. Individual carriers may modify this basic rule with FAA approval. For example, some carriers require a pretakeoff contamination check from the cabin, regardless of whether the holdover time has been exceeded, anytime precipitation has fallen after the beginning of the antiicing process.

Knowing that a visual check is required when using holdover times is one thing, but knowing what to look for is quite another. What does fluid look like when it has failed or is about to fail? Assuming that a pilot can actually see the fluid on the wing—a highly questionable assumption at night with scratched cabin windows and wing lights that may or may not be helpful— what a pilot should hope to see is a clean, uniformly shiny surface, whether it is tinted orange, amber, or green. It should be glossy. It should not have patches of discoloration or texture that would indicate "loss of gloss," which is what happens when the fluid has absorbed all the moisture it can and is beginning to fail and turn to slush. Above all, no snow should be beginning to accumulate on top of the fluid. Any of these conditions indicate fluid failure.

Even in the best of circumstances, fluid failure is often extremely difficult to determine; and if the deicing/antiicing fluid fails, the aircraft will not meet its certified performance and handling standards for takeoff. The results of a test program carried out in Canada during the 1996–97 winter season confirmed that, on a B-737, pilots did not recognize fluid failure during the visual inspection until 10 percent of the fluid on the wing had failed. When Type I fluid was used in the tests, the rapid propagation of fluid failure (typical of Type I) resulted in as much as 25 percent of the fluid failing before the pilots recognized the failure. Additionally, for the Type I tests, "wings were on average 40 percent failed five minutes after first failure…." This fact clearly raises safety concerns about the allowable 5-minute delay between the pretakeoff check and the takeoff if Type I fluid has been used for antiicing.

ALPA’s Ground Deicing Project Team believes that pilots must become more involved in making decisions about ground deicing/antiicing. All too often, ground personnel who have little understanding of holdover time and are motivated by concerns that are not relevant to the flight crew make decisions about the type of fluid used for antiicing. If freezing precipitation is falling at the time of deicing (or is forecast to occur before takeoff), Type II or Type IV fluids are the only appropriate fluids to be used for antiicing. Type I fluid is not appropriate for antiicing except in conditions of very light snow flurries.

Although ALPA recognizes published holdover times as "guidelines," pilots should be extremely cautious about exceeding them. Above all, when in doubt about whether you have a clean wing for takeoff, return for a reapplication of antiicing fluid.