The Next-Gen Aviation Weather Frontier

Airframe icing is insidious. For general aviation pilots, that means it proceeds in a gradual, subtle way, but with harmful effects.

Therefore, icing remains one of the most serious hazards for small and medium-sized aircraft—often invisible, sometimes unexpected, and always demanding proactive risk management even when the aircraft is equipped with a certified ice protection system. In the terminal environment—takeoff, climbout, approach, and landing—icing can escalate from nuisance to emergency in a matter of minutes due to its transitory nature. 

Recognizing this challenge, the Terminal Area Icing Weather Information for NextGen (TAIWIN) initiative is being spearheaded by the FAA to provide more precise diagnostic and forecast tools for icing conditions within the terminal area. 

Let’s take a look at TAIWIN with a focus on operational safety—what it is, why it is needed, how it works, and where it is heading.  

But before diving too deep into TAIWIN itself, it’s worth reviewing why icing in the terminal area deserves our full attention. Some of the most serious icing encounters will be when approaching or departing an airport, especially for those flying in IFR conditions. It is not uncommon for an aircraft to experience  a rapid onset of ice accretion within the terminal area. This region often has complex micro climates that include freezing drizzle (FZDZ), freezing rain (FZRA), and small-drop icing conditions where certification and operational rules may be restrictive. 

Many GA aircraft are not certified for flight into known icing (FIKI), especially supercooled large drop (SLD) conditions that occur with FZRA/FZDZ. Traditional weather briefings that include TAFs, METARs, SIGMETs, G-AIRMETs, and PIREPs provide important weather data but often not the detail pilots need in the airport’s terminal area. Enhanced icing awareness in this critical region means better planning (pre-takeoff), avoidance options (climb and diversion), and safer approaches and landings.

For general aviation pilots especially, where resources are more limited than with heavy iron, tools like TAIWIN promise a significant safety upside.

What Is TAIWIN and Why It Matters

So, what is TAIWIN? As mentioned, TAIWIN stands for Terminal Area Icing Weather Information for NextGen. In essence, it is an FAA research and development effort to create a high-resolution diagnostic and forecast tool specifically for icing conditions in the terminal area. That’s defined by the National Weather Service (NWS) as a region within 5 sm of the center of the airport’s runway complex. This is the region that forecasters use when issuing Terminal Aerodrome Forecasts (TAFs). 

For TAIWIN, however, the terminal area is much larger and is the region within a 30 nm radius around an airport, from the surface up to approximately 12,000 ft agl. In a recent post about TAIWIN, the FAA claimed that the “idea is to allow pilots to get ahead of the curve by getting the drop on icing conditions before icing can put them in peril.” In short, TAIWIN aims to fill a gap—a tool between broad-scale icing forecasts and the real world of approach and landing airframe ice hazard.

The focus of TAIWIN is to diagnose and forecast icing types including FZDZ, FZRA, and small-drop icing conditions distinct from the larger-scale en route icing tools such as G-AIRMETs and the Current and Forecast Icing Products found on the Aviation Weather Center’s website (aviationweather.gov). It will also capture when no icing is expected in the terminal area. The goal of TAIWIN is to provide timely, location specific icing information for flight planning (pre-departure) and tactical use (approach, climb, and landing) from the current time to 12 hours in the future for select terminal areas across the conterminous U.S. Keep in mind, TAIWIN is still a research tool—not yet operational—but undergoing user evaluations and algorithm development at this time.

How TAIWIN Works

TAIWIN ingests multiple data sources. Its diagnosis and forecast components both employ what is referred to as a time-lagged ensemble of model output from the High Resolution Rapid Refresh (HRRR) numerical weather prediction model. This produces the high-resolution grids of icing needed within the terminal area. The diagnostic combines these model grids with surface observations (e.g., ASOS and AWOS) and remote sensors, radar and satellite information of precipitation, clouds and drop size to produce a more comprehensive assessment of the icing environment.

Before this effort could move forward and to improve the understanding and prediction of aircraft icing, in early 2019, the FAA’s Aviation Weather Research Program (AWRP) sponsored the In-Cloud Icing and Large-Drop Experiment (ICICLE) field campaign. Using the National Research Council of Canada (NRC) Convair 580 aircraft and surface sensors at five Midwestern airports, researchers gathered about 120 flight hours of in-situ measurements in small drop, freezing drizzle, and freezing rain environments, as well as mixed phase and non-icing environments. 

The ICICLE findings are directly shaping TAIWIN’s diagnostic and forecasting algorithms, which continue to evolve through ongoing research. This dataset now serves as the backbone for TAIWIN’s validation efforts.

One of the problems that occurs when issuing forecasts of any type (e.g., icing, turbulence, etc.) is the seamless transition from observational guidance (what is happening now) to forecast guidance (what will happen in the near future). This can result in harsh transitions between observations and forecasts that could impact operational decisions. In an effort to make this as seamless as possible, the TAIWIN program is developing a nowcast component. This is accomplished through what is known as persistence forecasting. 

In forecasting, persistence is a very simple prediction method that assumes future values will be the same as the most recent observed value. Meteorologists at the 122 local Weather Forecast Offices (WFOs) situated around the country often use persistence as the method to predict the first few hours of a TAF. Therefore, the first forecast group in the TAF may remarkably look a lot like the latest surface observation (METAR) for the airport. You may have seen similar features used for weather radar tracking. 

Over the past several decades, radar feature tracking has shown strong skill in the short-term forecasting of deep, moist convective systems, motivating interest in applying similar techniques to the detection of icing-related features in radar and satellite data. Short-term forecasts of these icing-relevant features have demonstrated reasonable skill to two hours or more, providing developers with confidence that icing nowcasts could offer meaningful operational benefits for those pilots that use TAIWIN. 

Certification and the Road Ahead

TAIWIN is about improving safety, however, there is also some regulatory context needed. For an aircraft to be certified for flight into known or forecast icing conditions, it must be in accordance with 14 CFR Parts 23 and 25. These are the regulations for airworthiness standards. One of TAIWIN’s goals is to distinguish between small-drop icing, FZDZ, and FZRA that is expected to occur in the airport’s terminal area. That differentiation maps directly to 14 CFR Part 25 Appendix C versus Appendix O regimes. 

Appendix O and Appendix C refer to two different sets of icing certification requirements used by aviation authorities—primarily the FAA—to ensure that aircraft can safely operate in various in-flight icing environments. Appendix C represents the traditional icing envelope defined several decades ago, based on supercooled liquid water drops typical of stratiform-type cloud conditions. It specifies ranges of drop sizes, liquid water content, and temperatures in which aircraft must be able to safely fly, detect ice, and activate ice-protection systems. Most legacy aircraft are certified only to Appendix C, meaning they are proven safe in “small drop” icing but not necessarily in more complex atmospheric conditions that create an environment with larger drops.

Appendix O, introduced more recently, addresses a different and more hazardous type of icing: supercooled large drops (SLD), which include freezing drizzle and freezing rain. These drops are larger and can spread aft of protected surfaces, allowing ice to form in areas not covered by boots or a heated leading edge—potentially degrading lift and control more severely. Appendix O therefore includes requirements for aircraft to demonstrate safe handling, performance, and ice-protection capability in these SLD conditions, which often produce shapes and accretions not captured under Appendix C testing.

Aircraft certified for both Appendix C and Appendix O conditions are validated against a wider range of icing phenomena, improving operational safety—especially in regions or altitudes prone to mixed or complex icing. The FAA does not publish a list of specific models, as compliance is part of the individual type certification process for each manufacturer that is often kept under a tight lid. But, I do not believe there have been any aircraft certified under 14 CFR Part 25, Appendix O as of yet. This is because aircraft manufacturers face several significant challenges in complying with the rigorous 14 CFR Part 25, Appendix O icing certification requirements, primarily stemming from the complexity of SLD conditions and the difficulty in replicating them for testing.

For example, the NASA Glenn Icing Research Tunnel (IRT) in Cleveland—affectionately called the “ice house”—is where many manufacturers go to test their aircraft designs for Appendix C certification. The IRT is a closed loop with a 5,000 hp fan containing wooden blades made of spruce that can generate winds approaching 400 mph. Air in the tunnel can also be chilled to a frigid temperature of minus-40 degrees Celsius, which is plenty cold enough to test at all ranges of potential icing regimes. 

Even with these conditions, the IRT cannot replicate Appendix O conditions. Despite high wind speeds in the tunnel, many of the larger drops would succumb to gravity and fall out before reaching the model in the test section. What this means is that aircraft manufacturers will need to find alternative and likely more expensive solutions to certify the aircraft for flight into these conditions. In case you were wondering, large turbofan aircraft (e.g., Boeing 737) operate under an FAA waiver and have never been certified to fly into these large drop environments.  

So, what is the output of TAIWIN going to look like? According to the FAA, it will be using high-resolution, 3D-gridded fields showing probabilities of each icing category (e.g., FZRA, FZDZ, small-drop icing) in the terminal area from the surface to 12,000 feet agl). A simplified “one answer” output for each terminal area is expected where the algorithm determines the predominant icing category and assigns a single label to that area. The initial target is a product that gets updated every 15 minutes, plus a short-term forecast (e.g., up to 12 hours). 

SIGMETs for severe ice and urgent pilot weather reports will still be controlling no matter what TAIWIN suggests. The FAA announced that the development roadmap aims toward demonstration by 2028 and eventual operational deployment sometime thereafter. Recent FAA staff reductions and the government shutdown last year have had a major impact and created a significant backlog that is likely to extend this timeline. The development of TAIWIN represents a meaningful step forward in icing risk mitigation at these critical phases.

But remember: Safety in icing is never about one tool but rather about a system of awareness, planning, aircraft suitability, and decision making. TAIWIN is a promising addition to that system, but the pilot still remains the ultimate safeguard. Fly safe, watch your icing risks, and stay current.

This column first appeared in the March Issue 968 of the FLYING print edition.