Wx Watch: Internet Icing Upgrades

Marcia Politovich: The Icing Life

By Thomas A. Horne

Marcia Politovich is the deputy director of Science, Aviation Applications Program at the National Center for Atmospheric Research (NCAR). With a doctorate in atmospheric physics from the University of Wyoming, Politovich coordinates icing research at NCAR, conducts research in icing environmental characterization, has published numerous professional papers on aircraft icing, and was central in developing the Current Icing Product (CIP) models for the Aviation Weather Center's Aviation Digital Data Service (ADDS) Web site. She is also an advisor to the FAA's Inflight Icing Steering Committee.

AOPA Pilot: "How did you begin your involvement in aviation icing research?"

Politovich: "I was working in cloud physics for my master's degree at the University of Washington and they had this old Douglas B-23 that they did research in. They said, 'Why don't you come along on a field project in Montana? We're taking the B-23, and we could sure use another person on board.' It was a cloud-seeding project, and this airplane was instrumented with a lot of cloud-physics measuring instruments, and we were measuring the response of clouds to seeding. We had a number of airplanes on that project, and one was an Aero Commander. It seemed like every time it went into the clouds they had to go to a lower altitude, deice, and then come back up. We were in this B-23 and it could pretty much fly through the ocean, I think. We got iced up and it was no big deal. Just pop the boots and go through it again. So that was the first time I was really exposed to icing. I did get interested from that experience and I did more flying in the B-23. And then at Wyoming they had the [Beechcraft] Queen Air and the King Air, and I started working with Wayne Sand. So that's how I got interested in aviation weather."

Pilot: "What was your first icing research project?"

Politovich: "It was the winter of 1980-81. Wayne had a project to do some research for the FAA Tech Center to take another look at the icing certification envelopes. With modern equipment, was there anything missing from the envelopes? So we went around the country, and we compiled the numbers and we found that those guys kind of did it right back in the 1940s and '50s when they first made the envelopes."

Pilot: "Until large-droplet icing came along, right?"

Politovich: "In fact, one of the first modern encounters with large drops was with a program run by the University of Wyoming. They were working a weather mod project in the Sierra Nevadas, and they got into a situation where they weren't sure what was going on, but they were losing lift and airspeed really fast, and I heard it got quiet in a big hurry up in the cockpit."

Pilot: "What kind of airplane was this?"

Politovich: "A King Air 200T, and they came back and analyzed that data and found there were a lot of drizzle-size drops there. We reported on it and published a few papers and got some notice, but not 'gee, that's interesting.' Then when the Roselawn accident happened in 1994, one of the meteorologists at the NTSB gave me a call and said, 'The Roselawn thing sounds a lot like the stuff you guys published on large drops, so give me a call — we might work on this.' And that whole thing snowballed into a whole series of events by the FAA, and now it looks like they're going to take some certification action on it."

Pilot: "So who came up with the idea for the Current Icing Product (CIP)?"

Politovich: "It was our group at NCAR, I'd say. Ben Bernstein was a large part of it. You know, when you're a forecaster you consult a lot of resources. You don't just look at models, or satellites, or pireps. You put all that information together, and we thought, well, why don't we just try this with an automated program. Paul Schultz at NOAA and I had come up with some guidance from using numerical weather models at the holding temperatures in high relative humidity, but we thought, 'That's OK, but we can do a lot better if we pull in some other information.' So we worked on that and came up with what we called the 'stovepipe algorithm.' It was one of the first responses to the Roselawn crash. The reason it was called 'stovepipe' was because it looked for surface reports of freezing drizzle, and then we made it a little stovepipe that centered on that and looked up into the atmosphere, where you might have supercooled large droplets...."

Pilot: "How does the CIP differ from the stovepipe? What are the inputs?"

Politovich: "The CIP has more data. It includes the numerical model — the Rapid Update Cycle, or RUC — METARs, GOES satellite imagery, the satellite/radar mosaic, and pilot reports."

Pilot: "What are you looking for in the METARs?"

Politovich: "Precipitation type. And we're also doing something now with precipitation amount. We have a hierarchy of precipitation types. Freezing drizzle is the worst, followed by freezing rain, ice pellets, then snow, rain, and drizzle, in that order."

Pilot: "How do you verify the accuracy of the CIP?"

Politovich: "We use pireps, which sounds incestuous, but when we make the CIP we use pireps from an hour up to the issue time. And we verify by using pireps from the following hour. We also have access to research aircraft data from NASA Glenn and from the University of North Dakota."

Pilot: "Why did you change from the old CIP to the new CIP?"

Politovich: "The old product used to have something called a 'potential,' which was an uncalibrated probability. We have since calibrated that, and on the new product you will see that it is icing probability. It's the shades of blue, which are severity, with the overlaid SLD areas in red, and then you can use a pull-down menu and get different probability thresholds. Before, with the old product, it was 'you are in such-and-such chances of any kind of icing,' and now it's the probability of a certain intensity of icing, which is a lot more valuable.

"We also had to do the new CIP because of all the regulations for dispatching in icing conditions [ in Part 135 and 121 operations — Ed]. You know, 'Thou shalt not fly an uncertified airplane in moderate or greater icing.' So to be legal and to make this an unrestricted product we needed to show severity, and demonstrate that we can do that fairly well. At least as well as the airmets do. Which is one of the big changes with the new product; it's unrestricted. It's operational, which means anybody can use it. Which opens up its use in the cockpit. If weather vendors want to include it with their suite of products they can send up to the cockpit."

Pilot: "What's next in icing research?"

Politovich: "We're doing some things with improved radar detection of icing conditions. We're checking out this dual-polarmetric upgrade to Nexrad that's about to happen at the end of this decade. It gives you information about the shapes of things in the atmosphere. To tell if it's a round drop or a flat snowflake.

"With dual-polarmetric data we start identifying snow, snow mixed with rain, and freezing drizzle. This would be great information in the terminal area within a 50-km radius of an airport when you're bringing planes down through that stuff and holding them. We saw that problem at Roselawn, and where they were holding it was probably the worst place they could have chosen to hold. There may be things we can say based on dual-polarmetric information — 'Don't hold in this position,' or bring aircraft in on a little steeper trajectory so they miss the icing."

Pilot: "So your feeling is that the private vendors will be supplying the new CIP to the cockpit via datalink?"

Politovich: "I think XM satellite weather will be first, if they're not already doing it. And WSI will eventually be doing it." — TAH

WX BRIEFS

Web sites and self-briefing

In response to January's " Wx Watch: Icing, the Win a Six, and Expo," many readers requested a list of my top Internet weather sources.

Here's my list:

• AOPA Online Weather
• AOPA's Real-Time Flight Planner
• ADDS
• Storm Prediction Center
• NCAR
• Forecast models
• Hurricanes
• Soundings
• GOES (Geostationary Satellite) Server — TAH

Just in time for icing season, the Aviation Digital Data Service (ADDS) Web site has come up with a brand-new means for determining the probability and severity of an icing encounter.

The new tool is called the Current Icing Product (CIP), and it can be found under the Icing tab on the ADDS home page. After you click on the Icing tab, find the CIP information by clicking on the Supplementary Icing Information (CIP/FIP) field to take you to the data.

The new CIP essentially replaces the old CIP (the similar acronyms are confusing, I know), which stood for Current Icing Potential. The old CIP showed areas where icing could be anticipated, and expressed them in terms of probability. A color scale across the bottom of the map gave a color range, going from white (0-percent probability) to red (100 percent). As useful as this information was, it spoke only in terms of probability. There was no mention of the severity of the icing conditions depicted.

Would there be a 100-percent chance of risking a flight through light icing conditions, or heavy icing? There was no way of knowing — although in 100-percent areas you could make the reasonable assumption that the icing would be more severe.

The Current Icing Product (hereafter known as the "new CIP") addresses this shortcoming. It does this via four different images, selectable from a drop-down menu. These images have four different definitions, and here's what each shows:

• The Icing Severity + SLD product plots suspected areas of conventional icing and supercooled large-droplet (SLD — the kind of extra-large droplets that typically exist in drizzle conditions, and caused the much-documented crash of an ATR-72 commuter airliner at Roselawn, Indiana, in October 1994). Icing zones are depicted as ranging from Trace to Heavy. Using other drop-down menus, you can select from a range of cruise altitudes. Icing pireps are plotted as the familiar "pitchfork" symbols.
• The Icing Severity product shows all the above — minus SLD notations.
• The Icing Severity (prob > 25%) view plots suspected icing zones in shades of blue, as well as SLD conditions and pireps. SLD symbology is a stippled red, with a darker shade of red indicating a better than 50-percent chance of SLD. The blue zones are where the computer model estimates a greater than 25-percent chance of encountering icing. Using this view, you can quickly determine that your proposed route could have a better than 25-percent chance of icing, and that that icing could be light, moderate, or heavy. The new CIP, recognizing that different aircraft types vary in their ice-collecting efficiency, eschews the use of the term severe when it comes to icing intensity. After all, what's severe to a Mooney may be a passing annoyance to a Learjet. Gray shading identifies where icing probability is expected at less than 25 percent.
• The Icing Severity (prob > 50%) chart documents where there's a better than 50-percent chance of icing. Like the 25-percent probability charts, it also shows SLD areas and pireps, and uses gray shading to show where there's apt to be less than a 50-percent chance of icing.

Other products on the icing page will look familiar to repeat visitors to the ADDS site. For example, the composite, maximum icing potential for all altitudes can be selected (click on Icing Probability on the left drop-down menu, and Max on the altitude drop-down menu), and the Forecast Icing Potential (FIP) drop-down entries show projected icing for the next 12 hours.

These new CIP products, with their calibrated icing intensities, should prove extremely useful. Just remember that some icing events happen close to the ground or in highly localized form — like freezing rain — and may not be captured by the CIP algorithm. A lot can happen in the layers between the CIP's and FIP's 2,000-foot altitude intervals.

For all their added information, remember that the new CIP is a supplemental, not a primary, briefing source. Same thing with the FIP products. Airmets and sigmets remain the primary products for making go/no-go weather decisions. The FIP also should receive an upgrade to show icing intensities, and sources say that could happen as soon as this month. Stay tuned for updates, and get ready to see these CIP and FIP products added to datalink weather service packages in the very near future.

E-mail the author at [email protected].