Is that airframe icing or snow?

You’re flying in visible moisture with the air temperature below freezing and you notice something building up on the leading edge of the wings. Is that airframe icing?

What you see collecting on the wing leading edges in a cold cloud could be airframe icing, or it could be snow. Icing is bad, maybe very bad, but snow isn’t much of a problem. How do you know the difference?

What’s the difference between snow and ice on the wing?

Airframe icing, as I’m sure we all know, is created by super cooled moisture freezing as it strikes the airframe.

Super cooled sounds like an exotic term, but in this case “super” only means superior. In other words, “superior” to normal freezing, which happens at 0C or 32F. Super cooled moisture has been transported from above freezing air into below freezing air while remaining liquid, or at least not totally frozen.

Snow is frozen moisture. Snow can form as large or small flakes, even pellets, but by definition the moisture in snow is frozen.

How does airframe icing accumulate?

Generally speaking, super cooled moisture remains unfrozen in cold air because of surface tension on the moisture droplet. Under unusual conditions, much larger than normal super cooled drops can exist. This very dangerous condition is labeled on icing forecast maps as SLD for super cooled large drops.

Wing with snow
Is that ice or just snow?

What causes airframe icing to form is the impact of the super cooled droplet with the airframe disturbing the surface tension of the droplet. If the droplets are very small, the ice will form in rough layers called rime ice.

Rime ice does form on surface objects, such as trees or parked cars, as well as airplanes in flight. If the air and a surface, such as a tree branch, are both below freezing, super cooled droplets can freeze on the surface. Rime ice can build in thickness quickly because the freezing droplets pile up on each other in a rough layer.

The other broad category of airframe icing is clear ice, which is formed by larger droplets than cause rime icing. For clear ice to form, the super cooled droplet needs to be large enough for the liquid water to spread out before freezing solid. As the moisture in the droplets spread out and run back from the leading edge, clear ice is formed in a smooth, hard layer.

The third category of airframe icing—and the most common—is mixed icing. Mixed icing is simply a combination of clear and rime. With mixed icing there will be the rough, milky-colored rime, but also areas of clear, hard icing.

How does snow accumulate on the leading edge?

Because snow is frozen, it doesn’t stick to the airframe. Snow blows across the moving airplane just as it does across your car, or your driveway. The airframe is cold, below freezing, and so is the snow, so no melting or general adhering occurs.

But there is one important exception. And that is along the leading edges of your wings, and tail, and maybe the corners of the windshield. There snow can stick and form a line that you may think is airframe icing. But it’s not.

Snow sticking to the leading edge is sometimes called “impact snow” because those flakes have impacted the leading edge. That’s caused by the stagnation point.

Airflow over the wings and tail, logically enough, divides at the leading edges. But there is a small zone called the stagnation point where a small amount of the airflow impacts the leading edge instead of flowing over or under the surface.

As an aside, it is the existence of the stagnation point that allowed the brilliant Leonard Greene, founder of Safe Flight Instruments, to create the leading edge lift detector vane that is the stall warning system in hundreds of thousands of airplanes. In the 1940s, Greene realized the stagnation point on the leading edge moved up and down in direct correlation with angle of attack. His vane moves up and down with the stagnation point and sounds the alert when AOA nears the stalling angle.

It is along the stagnation point—really a small zone—that some snow flakes impact the leading edge and stick because of impact. The snow flakes can pile up to form a thin ridge that, at first, may look like airframe icing. In many airplanes there are other areas on the airframe, often on corners of the windshield, where there is a stagnation point and snow can stick.

Grumman Tiger with ice
Sometimes it’s easier to see ice on small objects like OAT gauges than wings.

Is it snow or airframe icing?

One sign of impact snow is color. The strip of snow will be white. That is particularly noticeable in airplanes with dark colored leading edges. Airframe icing will be either rough and milky colored if rime, or clear and smooth if clear icing.

Another way to know if you’re seeing snow or icing build up is to look at some visible small protrusion on the airframe. If you can see your OAT temperature probe, for example. Or the glare shields that many airplanes have to prevent the strobes and nav lights from reflecting back toward the cockpit.

A small object is an excellent icing accumulator, but not good at collecting impact snow. If you can see some small object and it’s not showing buildup, you’re looking at impact snow on the leading edges.

Will impact snow change to airframe icing?

Icing forecasts and airmets are particularly useless because if there is cold air and moisture, icing is always possible. So a forecast of a “chance” of icing is both accurate and useless.

If you’ve flown much IFR you have experienced a takeoff in rain followed by a climb into below freezing air and snow without any airframe ice at all during the liquid-to-frozen moisture transition. That seems impossible, but it’s really pretty common. The reason is the absence of a force to quickly lift liquid moisture into cold air without freezing it into snow.

Convective activity is a good lifting force to carry liquid moisture into cold air to create the threat of icing. You don’t need a thunderstorm to have enough convection to do the job. If the clouds are bumpy and look billowy before you fly into them, icing is probably more likely than snow.

There are also mechanical lifting forces, such as wind blowing over mountain ridges, that can carry liquid droplets up into below freezing air.

Another factor to consider—and this was a Richard Collins favorite—is the source of the moisture that is making the clouds. If you are flying ahead of a front, or on the south side of a low pressure system, the moisture is being recently ingested and could still be liquid. Behind the front the cold air has had time to freeze the moisture. To the north of a low the moisture has traveled a long way around the counterclockwise circulation and has had plenty of time to freeze.

Though airframe ice can form at any temperature below freezing, it’s uncommon at temperatures significantly below freezing. If it’s cold outside, and you see snowflakes whizzing past, you can be pretty sure that white stripe on the leading edge is impact snow, not ice.

And when it’s impact snow, it’s not a threat. The dynamics of the stagnation point will limit the size of the snow accumulation so it’s not a threat to disrupt airflow changing the performance of the airfoil or adding significant drag.

A white stripe on the leading edge will be alarming in a cold cloud, but before you head for the exits decide if it’s icing that requires immediate change in altitude or course, or impact snow that won’t matter at all.


  • Impact snow is not so bad for airframe icing but it does affect induction on a piston engine. I was flying through snow recently, no accumulation and my alternate air popped on.

    • You’re right, Ethan. The turbocharged Mooney 231 is the most alarming example of having the induction air filter icing over while flying in snow.
      The alternate air door in the 231 is located in a very warm spot in the cowling to prevent any further possibility of icing. So when the alternate air door pops open at a higher altitude the engine is sucking in warm less dense air, and the ram air from the normal induction and most of the turbo boost is lost causing an instant drop of many inches of manifold pressure depending on your altitude.
      Many pilots–including this one–were convinced the 231 engine had failed because the engine is suddenly running rough because the mixture is set at the pre-iced manifold pressure, and the MP has plunged. If you leaned to the “new” iced over manifold pressure the engine would run smoothly, but power output was greatly reduced.
      The first time it happened to many 231 pilots they didn’t diagnose the problem, but as they descended being certain the engine had failed, it would start to run smoothly at a lower altitude. Sometimes the ice on the air filter melted and allowed the alternate air door to close, or they descended to a point where the mixture now matched the ambient manifold pressure.
      After the first time or two the iced over air filter happened in the 231 you just adjusted the mixture for smooth operation and tried to find snow free air to get the cruise power back.
      Mac Mc

  • Mac does well to mention that there’s three temperatures to keep in mind.

    Everybody focuses on precipitation temperature, but airframe and air temperature also play roles.

    I’ve seen frost on the upper surface of a cold soaked fuel tank on a 20C summer day. Humidity can also play a role.

    One airline captain took umbrage when after boarding I asked if the drops on the wing were ice or water. He asserted that at 3C it was too warm, but I suggested it would be wise to check and he did get a rampie to go check.

    If you just came down from FL300 or so, drizzle will happily freeze to your airframe.

  • I well remember this conundrum from my turboprop days. Part of the issue is whether the snow is dry or wet. In September of this year, The NTSB issued a Safety Alert 082 addressing their concern with the icing effects of wet snow. The FAA has said in AC 91-74B that if “wet snow does begin to stick, it should then be treated as an icing encounter because ice may begin to form under this accumulation of snow.”. The Board is frustrated because the FAA has never defined wet snow, and they believe that pilots may have assumed that snow conditions “were too dry or cold to pose an icing hazard”.

    The effects, if any, of as-yet-undefined wet snow on airframe icing have never been investigated nor are they a class of conditions considered in icing certification. That said, it is not uncommon for a surface observation of snow to be associated with an icing accident; most likely this is the result of multiple precipitation types, but this is not always clear. Keep in mind that an ASOS LEDWI precipitation sensor can only report one type of precipitation, and the ASOS system will always give priority to a snow observation. It is only at larger airports with service level A, B or occasionally C (if the tower controller can leave the cab) that a human observer can augment a METAR and include multiple types of precipitation. So a METAR report of only snow, particularly at service level D (uncontrolled) airports, may very well be misleading.

    And it cannot be said enough that if there is any contamination of any dimension physically stuck to the wing, regardless of reported or observed precipitation types, it will result in some performance degradation.

    • Hi Charles,
      The only absolute I can say about airframe icing is that if there is visible moisture, and the air temperature is below freezing, icing is possible. Other than that, all we can discuss about icing in probabilities.
      And if it’s cold–say 20F or less–on the ground, and it’s snowing, the probabilities of finding icing above that spot are low. A strong inversion above the snow layer is possible, and it could support super cooled droplets, but that would be uncommon.
      What Richard Collins and I always discussed was having a way out of any icing you encounter. For example, if the ceiling is a couple thousand feet above the terrain, and you find ice and can’t hold altitude, you have an out of the clear air below the icing before you hit the ground.
      Or, if you have reliable reports of low tops that’s an out to get above icing quickly.
      A real challenging icing environment is pretty common over the great plains where there is a strong inversion with warm air aloft and a shallow layer freezing temps at the surface. Many times I’ve begun an approach in warm air and started collecting significant ice on the way down the glideslope. A climb back to warm air would shed the ice, but at some point either a diversion or a commitment to land are your only choices.
      Mac Mc

  • Reading the above prompted a thought; as a former working pilot and CFI-I, it concerns me that most of the articles I read on the subject of airframe icing in General Aviation aircraft with limited climb and altitude capability do not mention a primary method of dealing with icing.

    My preferred first avenue of retreat when encountering airframe icing in such aircraft is to start or continue climbing immediately (while it’s still possible) in the hope/expectation of reaching non-icing conditions above your present altitude.

    The decision to climb IMMEDIATELY allows for a greater choice of options if you do not succeed in climbing out of the icing zone. Often the accumulated ice will sublimate before it is time to land with a compromised airfoil.

    Consider the beginning of any significant accumulation of airframe ice in an aircraft with limited climb capability an emergency, in order to avoid an actual emergency. ATC will be happy to accommodate you without the need to declare an emergency.

    Background: 8 years and 3,000+ hours flying a normally aspirated BE55 (turbocharging would have been a huge safety factor) 5-6 days per week, encountered lots of ice under varying conditions. The aircraft had boots, and alcohol props and windshield. It was NOT certificated for flight into known icing for very good reasons – the alcohol system often failed (ran out of alcohol because of poor system design and implementation) and in severe icing, as is often encountered over the Rocky Mountains, the boots could be overwhelmed. I used Icex and a portable O2 bottle.

  • Thanks for the response Mac. That’s the way I’ve been rolling in my non deiced A36. In fact cancelled a trip to Danbury last Sunday. I had a plan for the ice, but no plan for the 1/2 mile vis in moderate snow. Take care.

Leave a Reply

Your email address will not be published. Required fields are marked *