Several months ago, I built an ADS-B receiver that pings planes flying around NYC, mainly to/from LGA due to my location in the northwest part of Manhattan. This receiver then feeds the data to the servers of FlightRadar24, PlaneFinder, and a nonprofit called OpenSky that gives flight data to academics for use in their papers. I originally built this receiver to gain cheap access to a number of flight-tracking APIs, because I wanted to pull data related to a test concept called ‘Wake Energy Retrieval’ that is being considered for widespread use in commercial aviation.
Consider the iconic flock of geese, heading south for the winter, in a classic V-formation. Why do geese instinctively make this formation? It isn’t to ‘draft’, at least not in the sense of reducing wind resistance. The largest benefit of the V-formation for a trailing goose is actually the uplift generated from the wingtips of the goose ahead of it. As the wingtips of the leading goose slice through the air, they generate a spiral vortex of air, similar to the sort that sharklet wingtips are designed to reduce on airplanes. Riding that wake in the right spot can generate an updraft that makes it easier for the goose to stay comfortably aloft, which is a major concern when it takes you a week to fly from Nunavut to Alabama.
Hence, biomimicry: what if airplanes could do the same? It turns out that with the right tools, they can, which is the concept behind ‘Wake Energy Retrieval’ initiatives. Wake Energy Retrieval is a fancy term for what geese do, which is wake-surfing, and the advent of widespread use of ADS-B signaling technology — automatic signals sent out by commercial aircraft at a 1090 MHz frequency — has made it possible for commercial airliners to do the sort of precise navigation necessary for WER.
The most recent trial of WER in commercial aviation took place on November 9, 2021, when Airbus — in coordination with SAS + Frenchbee on the airline side, and a variety of Atlantic-adjacent national agencies on the air traffic control side—successfully drafted an A350 test aircraft behind another A350 test aircraft on a flight from Toulouse to Montreal:
Using the Playback feature in FlightRadar24, I rewound to 9/11/2021 (European date formatting to honor the European company that executed this test), and tracked the flights by filtering to the AIB callsign, which indicates Airbus company aircraft. You can see the 2 A350 test aircraft — callsigns AIB01 and AIB02 — overlaid essentially on top of each other as they exit the Bay of Biscayne into the wider Atlantic, not long after departing Toulouse. AIB01 appears to be ahead, which means that AIB02 might be the trailing — therefore, wake-surfing — aircraft.
But then, at approximately 9:17:00 UTC — shortly after the 2 aircraft drop off consistent contact with ground-based ADS-B receivers, demonstrated by the black dotted line that indicates FR24 is estimating rather than actively tracking the aircraft position — the lead AIB01 appears to stop moving, while AIB02 continues to fly and transitions to satellite-based ADS-B tracking (indicated by the transition of the aircraft’s color to blue rather than the typical yellow).
6 minutes later at 09:23:00 UTC, AIB01 disappears from FR24’s ADS-B coverage entirely — the program is unable to continue playback while that flight is selected.
Meanwhile, however, AIB02 continues its flight, charting a direct course over the Atlantic rather than adhering to the North Atlantic Track routes — the colored lines that span the North Atlantic in the above screenshot.
In fact, re: the NATs, the choice of venue for this test flight— the North Atlantic — was deliberate. Experts view the North Atlantic as the perfect trial ground for WER, due to the unique way that transatlantic airplane flows are managed. Because the skies over the North Atlantic are so crowded, but are nonetheless beyond the reach of traditional radar-based airplane tracking systems, a system known by multiple names but commonly called the North Atlantic Tracks was developed back in 1961 (a similar maritime system was developed even earlier in 1898). Using jet stream measurements, weather forecasts, and other factors, the ATC centers at Gander in Canada and Shanwick in the UK coordinate to publish ~8 tracks upon which most trans-Atlantic flights will chart a path for much of their flight time over the central Atlantic. A similar, albeit less crowded, system governs trans-Pacific flights.
Essentially, the North Atlantic Tracks are giant sky highways, indicated by the colored lines in the above screenshot and also over the Pacific in the below screenshot:
The NATs funnel many flights into multiple uni-directional streams — the perfect environment for convenient WER pairings.
At approximately 12:47:13 UTC, as AIB02 comes within range of standard ADS-B tracking technology off the coast of Newfoundland (rather than the satellite-based ADS-B used over the central Atlantic), AIB01 winks back into existence — still right ahead of AIB01. What happened to this ghost flight over those ~3 hours when FR24 wasn’t capturing its flight, or even estimating its flight path? Was a single satellite link sufficient to track both of the flights, given that they were presumably drafting as essentially 1 unit for much or all of that time? Why did that satellite link follow what appears to be the trailing aircraft, rather than the leading aircraft?
Regardless, AIB01 and AIB02 soon separate to a more typical distance as they approach their destination, and land in Montreal at roughly half past 3 PM.
Airbus is not the 1st manufacturer to perform such a test flight. Boeing performed a similar test back in 2018 with 2 777 aircraft, though they published far less data on the test results than Airbus. Moreover, military aircraft practice wingtip flying all the time. The concept is not new — but the technology necessary to make it commercially applicable has emerged from infancy, suggesting that primetime for WER might be approaching soon.
With technology out of the way, what are the 3 bigger hurdles? Regulatory regimes, operational difficulties, and commercial considerations. From a regulatory perspective, the level of schedule coordination necessary for commercial airlines to practice WER would be illegal, outside of the couple trans-Atlantic joint ventures that might be able to achieve the necessary scheduling.
But even harder than changing a regulatory regime would be the operational and commercial constraints. Consider the difficulties that many airlines worldwide have with operational metrics, at even the best of times! The idea that resource-constrained carriers would be able to reliably send out 2 aircraft from 2 different airports with sufficient accuracy that they could rendezvous in a single spot over the North Atlantic — reliably enough to make worthwhile the considerable investment into technology and training that will be necessary to achieve WER — is almost laughable. Moreover, would airline planners truly be willing to sacrifice the commercial benefit of optimal departure times, reorienting possibly their entire trans-Atlantic schedules around the chance to intercept their own (or JV partners’) aircraft and maybe glean a 10% fuel savings? It seems unlikely.
All these factors considered, most folks will conclude that there is only 1 feasible solution: the practice will have to be regulated, mandated, and coordinated by government air traffic control agencies. If an external body is responsible for pairing aircraft, airlines can continue business as usual without change to antitrust regulations/operations/commercial planning, while still being confident that if their A321LR happens to fall into line behind an A380 from a random foreign carrier with whom they have no relationship, an ATC body will organize a WER pairing and help the trailing aircraft save fuel. No need for a complicated joint venture, operational coordination, or shuffling around a schedule for an inferior flight time. A government-mandated solution — with input from both North American and European government agencies — will maximize the number of pairings (and therefore the amount of fuel saved + emissions reduced) while minimizing the possible disruptions or adjustments to commercial service.
Hence, I built the ADS-B receiver in order to gain access to the free Business/Premium subscriptions that come with feeding data to FlightRadar24 and Planefinder, which I hoped would help expedite access to the historical datasets that would allow me to programmatically gather data back to November when the Airbus WER test flight took place (rather than just manually using FR24’s Playback tool). OpenSky didn’t give me anything, but their mission is admirable (and their upcoming research symposium is on November 10th).
However — shortly after buying the Raspberry Pi (backordered because of the chip shortage) and the 1090 MHz SDR-antenna, I read on the NATS-UK blog that the North Atlantic Tracks are broadly on their way out. At the height of the pandemic, the low volume of trans-Atlantic flights allowed ATC authorities to test an exciting new concept: OTS-Nil. OTS-Nil days (OTS — Organized Track Structure — is another name for the NATS system) were 20 days in 2020 when no NATs were published at all. Moreover, as of March 2022, aircraft flying below 33,000ft are permitted to chart their own courses. These changes hint at a near future when, instead of being routed into narrow highways over the oceans, all aircraft will chart the most efficient courses to/from each city-pair across the Atlantic and Pacific. But in my estimation, those narrow highways were what really even made trans-Atlantic commercial use of WER worthwhile, as it funneled planes ahead/behind each other in a way that made wake surfing feasible.
And that means that widespread commercial use of WER becomes less compelling, perhaps fatally so. The removal of the NATs probably renders WER redundant, because airplanes will save far more fuel with a 100% chance to fly in a direct course from A to B, versus being routed into a narrow, inefficient flight corridor where they have a significantly <100% chance of pairing with a leading aircraft and engaging in WER for part of their journey.
So, while the concept of surfing a plane on another plane’s wake is inherently fascinating regardless of its commercial applications (and there may still be WER applications in other flight geographies), the lessened likelihood of widespread commercial trans-Atlantic WER immediately made the Airbus test flight data just a bit less compelling. But I had already bought the Raspberry Pi and the 1090 MHz SDR-antenna, and there’s not much you can do with a 1090 MHz SDR-antenna except to track planes. So I built the receiver anyway.
Would I recommend ADS-B receiver construction to others, even without a WER angle? Definitely. It’s a fun project if A. you’re more interested in airplanes than the average person and B. you want to mess around with a micro-computer. Other folks feed their data to FlightAware and ADS-B Exchange, as well. Regardless of which flight-tracking service you choose to feed, just make sure you give back — consider feeding your data to OpenSky!
The following links are excellent general-purpose resources for those interested:
OpenSky instructions:
https://opensky-network.org/community/projects/30-dump1090-feeder
FR24 instructions:
https://www.flightradar24.com/share-your-data
Planefinder instructions:
https://planefinder.net/coverage/client
An excellent FR24 forum guide on how to feed data to multiple sites:
https://forum.flightradar24.com/forum/radar-forums/flightradar24-feeding-data-to-flightradar24/10903-how-to-feed-data-to-multiple-sites-a-brief-guide