The problem with lost com procedures

I was working a flight from Houston to Newark Liberty International Airport in New Jersey—the airport most in the news lately for delays and equipment failures—when we got a message from our dispatcher to divert to Indianapolis International Airport because of Newark’s radar failure. This led to some back-and-forth confusion, because not a minute later we got another message telling us to press on to Newark. That was followed up with a second request to divert, and this one was for real.

When I got on the ground in Indianapolis, I called my dispatcher to get more information. Apparently, Newark’s radar and radios had all failed, and when the radar was brought back online, only some of the radio frequencies were fully functional. The decisions made by the FAA (move to a slower arrival rate and partial ground stop to take into account fewer functional sectors) and the airline (use an internal plan for strategically diverting aircraft to minimize further congestion later in the day) meant that the several dozen diverted flights would be grounded for a while.

Over the next few weeks, there were more of these temporary outages, most lasting less than a couple of minutes, and most in the Newark airspace; one occurred in Denver, but was also short-lived. Without getting into the specifics of how and why the equipment failed when it did, this is a great opportunity to review how pilots are expected to handle a lost communication situation.

When a pilot on an instrument flight plan loses radio contact with the ground in visual conditions, the expectation is that the pilot will land as soon as practical. Pretty simple. When flying in instrument conditions, it gets more complicated—quickly. The Aeronautical Information Manual spells out the two-way communications failure procedures in Chapter 6, Section 4. Section 6-4-1 explains how the pilot is to determine the route and altitude to fly, as well as the timing of starting an approach. But the first sentence of 6-4-1(a) says it all: “It is virtually impossible to provide regulations and procedures applicable to all possible situations associated with two-way radio communications failure.”

In light of what happened at Newark on multiple days, it is clear the FAA wrote AIM 6-4-1 from the perspective of a single airplane losing radio communication, likely because of an equipment failure, bird strike, lightning strike, or some other problem. What is also clear is that no real consideration was given to a problem on the ground affecting a large swath of airspace crowded by dozens of aircraft flying at various speeds and altitudes. Further, no process is laid out by the FAA to help us determine when we are, in fact, NORDO (no radio). Any instrument-rated pilot can tell stories about being in a dead zone right over an antenna, or waiting for an air traffic controller change, or dealing with a controller who is working multiple frequencies. These are everyday events that rarely lead to more than a few seconds of radio silence from the ground.

But, how are we as pilots supposed to know when we have crossed the line into true NORDO or lost radar operations? And when do we initiate lost com procedures? Why does it matter? In a situation like this, the starting line hasn’t been defined. It isn’t a matter of not defining it well, but that it hasn’t been defined at all. In busy airspace like the New York area, the FAA has, over time, rigidly defined the boundaries of the major airport, each with its own approach control and each with its own departure procedures, standard terminal arrival routes, transitions, and separation procedures. Some secondary towered airports are just as busy, and there are myriad nontowered general aviation facilities.

The Newark terminal radar approach control facility's issues drive home the point that we are totally unprepared for a ground failure that could lead to a massive lost com situation. How long do we wait before we begin implementing NORDO procedures? There is a huge difference in the distance covered over time for an airplane at 10,000 feet flying 300 knots versus traffic below 10,000 feet that might be doing 250, or 210, or 180, or 110 knots. The procedures tell us to use the estimated time of arrival on our flight plans to try to time an arrival or an approach. Altitudes are based on published minimum altitudes, but we as pilots may not be aware of all possible altitude constraints, and there is even an example provided where a pilot in a terminal area needs to climb, possibly after descending, which is counterintuitive and workload intensive.

Another major issue is that ATC almost never has an airplane fly the actual filed course all the way to the runway. Vectors, headings, shortcuts, and not-so-shortcuts are common. If a facility goes down, how are we to proceed? How do we do it safely? What if multiple aircraft have the same ETA on their flight plan for the initial approach fix? We have no way of knowing any of this, and as reliable as the traffic alert and collision avoidance system is, I don’t want to rely on that as a way to determine how to navigate to a runway.

None of this takes into account any issues with weather. ATC and pilots have a different set of priorities when it comes to weather, and individual controllers and pilots can have different interpretations of the same weather. Search LiveATC.net for a busy Class B area on a bad weather day and you’ll hear controlled chaos, speech at a pace that even the best auctioneer would envy. Now imagine if all of that went…quiet. Who does what? When? How?

My intention is not to sow fear or even doubt. But the issue of how to cope with a ground failure versus a single aircraft issue is one that the FAA and industry need to address. Other major metropolitan areas have similar congestion issues: Central Florida and South Florida; Northern and Southern California; Dallas and Houston in Texas; the Baltimore/Washington, D.C., area, which also encompasses flight restricted zones and a prohibited area; and Chicago. Denver and Salt Lake City are also busy, with the added wrinkle of serious terrain.

I agree with the FAA that not every scenario can be imagined. But a loss of equipment on the ground because of a storm, an earthquake, a power failure, a larger grid failure, or even terrorism can’t be ignored any longer. The FAA is currently working with pilot unions, airlines, and others to try to craft a process for something that goes far beyond an airplane losing an antenna to a bird strike. This is an urgent priority, and it must be clear what pilots will be expected to do— in what will likely be one of the most stressful moments of their lives—without having to dig around for a long document with instructions.

So, what can you do?

The obvious question in all of this is: What can pilots do if they have reason to suspect a large-scale equipment failure on the ground? There is no great answer, and relatively few good ones. These are merely suggestions, but they might help spark some useful conversations. All of these start with the assumption that you are on top of your fuel situation in terms of both range and time.

Try 121.5 MHz, the emergency frequency. This is especially apt if you are in a relatively isolated area, or don’t have other frequencies immediately available. If something is indeed amiss, airliners and corporate aircraft will likely be on this international distress frequency, and your suspicions may be validated. Further, aircraft monitoring 121.5 may be able to suggest another frequency for you to try.

Try a nearby control tower, and if necessary, call whatever frequencies are listed: ground, clearance delivery, tower, and so on. If nobody answers and you have no reason to believe the facility is closed, you may be one step closer to confirming your suspicions. You may even get a local FBO on the unicom that can make a phone call.

Workload permitting (a huge caveat), try to contact flight service or anyone else who may be monitoring via a VHF omnidirectional radio range navaid (VOR). If you were never exposed to this as a student or just haven’t done it in so long you can’t recall the details, sit down with a CFI and review the steps, including how to determine what VORs are available. Then, get in the airplane and go do it, because it’s a handful—but it can work. I did this with all my students when I was a full-time CFI, and we did it on a cross-country flight to make it more realistic. In instrument conditions, doing this without an autopilot can quickly become overwhelming.

If you have an ADS-B display and you are in instrument conditions, and you are absolutely convinced that there is a widespread equipment failure on the ground, your fallback is your emergency pilot-in-command authority. Getting away from traffic is key. If you have any up-to-date weather information for any nearby airports that have an instrument approach or VFR weather, consider using that to get on the ground. Realize that this isn’t what the AIM says or what ATC expects, but it may be the safer option assuming you can expect the approach to be successful, there is no convective or icing activity, the runway is open, and other factors align. As soon as you land, you will need to get on the phone to flight service or whatever facility you can reach and make sure that they understand you have exercised your emergency authority and have safely landed. Write down the phone number you called, whom you spoke to, and the time and details of the call.

File a NASA Aviation Safety Reporting System report immediately.