In-Flight Icing

In-Flight Icing (IFI) continues to be a safety issue for aviation as it can distort the flow of air over the wing and adversely affect handling qualities. An airplane may stall at much higher speeds and lower angles of attack than normal. It can roll or pitch uncontrollably, and recovery may be impossible.

Damaged plane
Aircraft after being unable to climb to 5,000 feet. 'We were starting to pick up moderate rime icing, but the boots were still shedding the ice.'

National Transportation Safety Board findings showed that during the period 2008–2021, there were at least 52 aircraft accidents and 64 fatalities that identified structural, in-flight icing as a cause or factor. Several other recent cases are under investigation that could increase these numbers.

Collage of images showing ice on vairous parts of planes
Ice accumulation on National Research Council of Canada (NRC) Convair-580 during In-Cloud Icing and Large-drop Experiment (ICICLE) Field Campaign.

In addition, regulatory and policy changes have resulted in a shift in IFI research strategy. Traditional IFI forecasts focused on the probability and intensity of icing ranging from trace, light, moderate, to heavy. FAA policy changes in Title 14, Code of Federal Regulations Part 25 [Airworthiness Standards: Transport Category Aircraft], Appendix C, and Appendix O have been created to improve the safety in in-flight icing and super-cooled large drop (SLD) conditions.

These regulatory and policy changes have resulted in the following new icing categories for aircraft certification:

  • No cloud
  • Cloud but no icing
  • Appendix C icing conditions with maximum drop diameters (Dmax) < 100 microns
  • Appendix O icing conditions with Dmax 100–500 microns (Freezing Drizzle)
  • Appendix O icing conditions with Dmax > 500 microns (Freezing Rain)

To mitigate the in-flight icing safety hazard, the IFI research is focused on enhancing automated diagnostic and forecast capabilities used by pilots, dispatchers, and meteorologists. These enhancements allow for timely decisions on icing threat areas, optimum routings, and areas to avoid in compliance with updated regulations and policy changes.

Aircraft wing with ice viewed from aircraft window
Appendix O icing conditions running back on wing surface of an aircraft.

IFI Capabilities

The current IFI related operational capabilities are:

  • Current Icing Product (CIP) — an hourly diagnostics tool that makes use of satellite imagery, numerical weather prediction (NWP) model output, weather radar, lightning data, surface weather observations, and pilot reports by combining individual data sources into an integrated algorithm that makes use of the strengths of each capability while simultaneously minimizing individual weaknesses.
  • Forecast Icing Product (FIP) — a forecast version of CIP designed to mimic the technique of combining information from various sources and provides forecasts out to 18 hours.

The IFI program is focused on improving CIP and FIP by incorporating enhancements in NWP, weather satellites, and weather radar, and developing enhancements for Appendix C and O guidance.

The NWP model output will transition to the Rapid Refresh Forecast System (RRFS) model that has a horizontal grid spacing of 3 kilometers with 65 vertical levels.

Rapid Refresh screenshot on left, HRRR screenshot on right
13 km Rapid Refresh (RAP) horizontal grid spacing vs. 3 km High-Resolution Rapid Refresh

The newest generation of Geostationary Operational Environmental Satellites (GOES) provides an opportunity to enhance real-time detection of aircraft icing hazards. CIP will also be modified to incorporate Next Generation Weather Radar (NEXRAD) dual-polarization data from the Multi-Radar Multi-Sensor (MRMS) national mosaic. Research will also be conducted into the ability to diagnose inflight icing conditions based on Appendix C and O drop size.

CIP/FIP v1.x CIP/FIP v2.x
Inputs from NWP model and observations Inputs from NWP model and observations
Derivation of icing conditions largely based on interpretation of meteorological scenario Explicit use of model microphysics fields plus meteorological scenario analyses
Rules-based algorithms applied Rules-based algorithms; adapted for High-Resolution Rapid Refresh (RRFS) fields
Information blended with fuzzy logic methodology Information blending
Output products are icing probability, icing severity category, SLD potential Icing probability and severity plus drop size classification product
  Ensemble-based methods for ingesting model forecasts

CIP and FIP enhancements (Version 2.x) are targeted for initial NWS operational implementation in FY2025 and will include finer horizontal and vertical resolution and the use of additional weather radar and satellite information. Future versions will provide drop size information in accordance with the aircraft certification criteria. Development of this capability is dependent upon a full and complete analysis of the ICICLE field campaign data, which provides valuable in-cloud measurements of in-flight icing parameters. Future versions will add other refinements for improved accuracy and applicability to emerging aviation users.

product comparison screenshots
Current icing severity CIP on left and possible new icing category product (Appendix C and O) on the right.

Weather in Alaska presents a special challenge for icing detection and forecasting due to terrain, air-sea interaction, variation in weather across the state, and the lack of observations. An IFI product, Icing Product Alaska (IPA), was developed to provide experimental guidance for Alaska and surrounding areas. IPA was developed using the CIP and FIP as a foundation. With the implementation of CIP and FIP v2.X, output will be available over a large North American domain that includes CONUS, Alaska and off-shore areas. The experimental IPA will then be retired.

Last updated: Tuesday, March 26, 2024