D.C. Circuit May Blow Up the Remote Identification Rule for Drones
Lawfare contributors Ashley Deeks and Russell Spivak discussed, in 2017, the U.S. Court of Appeals for the D.C. Circuit ruling in Taylor v. Huerta that struck down the 2015 Federal Aviation Administration (FAA) interim drone registration rule. Some of the same players involved in that case, including Taylor’s attorneys, Jonathan Rupprecht and Kathleen Yodice, as well as the presiding judge, Robert L. Wilkins, reconvened on Dec. 15 for oral arguments in the case RaceDayQuads (RDQ) v. FAA, challenging the FAA’s remote identification (RID) rule. This post provides an overview of the rule, summarizes the challenges and offers analysis of the arguments as well as thoughts on what might happen going forward. Given that the FAA has made RID a prerequisite for certain operations over people, beyond visual-line-of-sight drone operations and the future unmanned traffic management system, this case has the potential to set the commercial drone industry back years and hamper national security agencies in protecting against rogue drones.
The FAA’s stated purpose of the RID rule is to protect airspace safety and national security by requiring all small drones (0.55-55 lbs) in the U.S. to broadcast out a “digital license plate” that is accessible in near real time to the FAA, national security agencies, law enforcement entities, other government officials and the general public.
It took the FAA more than two years to finalize the rule. The process began on New Year’s Eve 2019 when the agency published its initial Notice of Public Rulemaking (NPRM). Despite multiple requests to extend the deadline, the FAA closed out the public comment period on March 2, 2020. Even so, during that 60-day period, the FAA received more than 53,000 comments. Almost a year later, in January 2021, the FAA promulgated the final RID rule (which added a new Subpart C to Part 89 in Title 14 of the Code of Federal Regulations, Remote Identification of Unmanned Aircraft).
Under this rule, a drone will need to broadcast message elements (MEs) over WiFi and/or Bluetooth, including its unique identifier (e.g., a serial number), precise coordinates, geometric altitude, velocity, the control station coordinates or broadcast takeoff location and geometric altitude. In some cases, it will need to broadcast a time mark and an emergency status indication.
With regard to the range for broadcasting MEs, 14 USC § 310(g)(2) states:
Any broadcasting device used to meet the requirements of this section must be integrated into the unmanned aircraft without modification to its authorized radio frequency parameters and designed to maximize the range at which the broadcast can be received, while complying with 47 CFR part 15 and any other applicable laws in effect as of the date the declaration of compliance is submitted to the FAA for acceptance. (Emphasis added.)
Utilizing the 1 watt of output power provided for in FCC regulations, 47 CFR 15.247(b)(1), a RID system at +30 dBm (1 watt) on 2.4 GHz would transmit identification information more than 1 mile omnidirectionally. This will enable a wide swath of people, over large distances, to persistently track drones.
RDQ attorney Rupprecht performed a propagation study using this RID standard and compared it to the PCS 700 MHz band at +27 dBm, the cell phone addressed in U.S. v. Carpenter, covered previously in Lawfare. The blue area depicts RID and the smaller red circle, the Carpenter cell propagation (see Figure 1).
Figure 1. Depiction of the comparative reach of RID capabilities (blue) vs. the cell phone in Carpenter (red).
Source: Rupprecht Propagation Study Utilizing Google Earth.
This RID capability must be either hardwired into the drone (Standard Remote ID) or attached externally in the form of a module (Broadcast Module RID or BMID). Drones without RID capabilities may fly only in FAA-recognized identification areas (FRIAs), specially designated flight areas carved out for non-BMID or Standard RID drones, under the general purview of community-based organizations and educational institutions.
Manufacturers have until September 2022 to either integrate RID into their drones or build BMID modules. Drone operators have until September 2023 to either fly RID-compliant Standard or BMID drones or be relegated to flying in a FRIA.
When RID capabilities go live, not only will law enforcement and national security agencies be able to access a drone’s MEs in near real time, but they will be able to correlate the ME information with other personally identifiable pilot data in FAA databases. In other words, the rule enables persistent surveillance of drones and their operators in real time, with precise locational data, without a warrant.
This last bit caught the attention of a lot of folks, including some of the public commenters during the pendency of the rule. Another attention-grabber: The final rule deviated significantly from the FAA’s NPRM version, which had focused on a network-based RID solution. Under the originally proposed rule, Standard RID drones would have been required to have both network and broadcast capabilities. “Limited RID drones,” which did not make their way into the final rule, would have been network only. The final rule deviated from the proposed rule in other ways as well. Enter: RDQ.
RDQ is an Orlando, Florida-based multimillion-dollar e-commerce shop that caters to the first-person view (FPV) drone-racing crowd. FPV drone pilots, some of whom are children, use small, light drones to fly fast through elaborate tracks, in places such as parks and forests. Flying in a FRIA would be too constraining because these will not be the same type of public and wide open spaces that FPV pilots have the freedom to access now. And then there’s the geolocational aspect of RID, which could allow anyone, for example, to track and find a child flying a drone alone in the woods.
RDQ’s co-founder and CEO, Tyler Brennan, an active-duty U.S. Air Force F-15E pilot said of the suit that he seeks “to protect the constitutional rights of U.S. citizens to be free from unreasonable searches from the government when they are flying in their own backyards.”
To this end, Brennan gathered a legal team that includes not only two lawyer-pilots, Rupprecht and Yodice, but also a partner in the Parlatore Law Group, Elizabeth Candelario.
The RDQ team filed its case a year after the final RID rule launched. It was not until this past August, however, when the team filed its 14,000+-word brief and appendix, that the constitutional and other bases for its challenges became clear. RDQ alleged that the rule is a violation of the Fourth Amendment because it allows warrantless tracking in a backyard.
RDQ’s other challenges to the rule included that the FAA arbitrarily and capriciously relied on undisclosed ex parte communications during the rulemaking process, the final rule was not a logical outgrowth from the NPRM, the FAA failed to comply with a legal mandate to consult with Radio Technical Commission for Aeronautics (RTCA) and the National Institute of Standards and Technology (NIST), and the agency failed to address significant public comments as required by the Administrative Procedure Act.
After receiving a deadline extension, the government’s nine-person legal team fired back in October. John E. Putnam, the Department of Transportation acting general counsel and deputy general counsel, leads this group of attorneys from the Transportation Department, the FAA and the Department of Justice. The other members include three attorneys from the Justice Department, including Casen B. Ross, who argued the case on behalf of the government, and an FAA senior attorney, who was directly involved in its RID rulemaking efforts.
The government’s brief outlined a host of justifications and fallbacks to support the rule, some of which were never mentioned in the rulemaking. On the constitutional front, citing the Supreme Court case U.S. v. Karo, the brief took the position that merely requiring RID technology on board a drone does not equate to an unreasonable search.
The government also explained that the movement of planes in public view does not give rise to a reasonable expectation of privacy. As a failsafe, the brief also noted that even if the rule violated the Fourth Amendment, the special needs exception would legally justify it. The special needs doctrine recognizes that law enforcement’s “special needs” justify a search predicated on less than the Fourth Amendment probable cause requirement when the objective serves “non-law enforcement ends.”
In mid-October, the Association for Unmanned Vehicle Systems International (AUVSI), the world’s largest nonprofit advocacy organization for unmanned vehicles, filed an amicus curiae brief in support of the FAA’s position. In an interesting twist, in the context of disputing RDQ’s standing, attorneys for AUVSI, not the FAA or RDQ, opened up Pandora’s box on the controversial issue of “avigation,” or landowner rights in low-level airspace above their property. AUVSI averred that RDQ’s Fourth Amendment assertions must fail because the FAA has jurisdiction down to the lowest blade of grass, including in one’s own backyard, as I have explored previously.
Because of AUVSI’s brief, RDQ responded quoting the Supreme Court’s Causby case (“The landowner owns at least as much of the space above the ground as he can occupy or use in connection with the land.”) and the Department of Transportation’s own position in a September 2020 Government Accountability Office report that disavowed the FAA’s jurisdictional claims in the airspace over private property. In other words, even the FAA’s parent agency is on record directly contradicting the posture the aviation regulator has taken in this case on the extent of its authority.
On Dec. 15, Rupprecht argued for RDQ before a panel of three judges: Wilkins, Cornelia Pillard and Justin R. Walker.
Going right for the jugular on the Fourth Amendment, Rupprecht said, “People have a reasonable expectation of privacy in their backyard. … With RID here, law enforcement, without any warrant, and it’s unlimited, can track an individual in his backyard.”
The judges cut off Ross, the Justice Department attorney representing the FAA, after his opening sentence and spent the next 27 minutes peppering him with Fourth Amendment-related questions about reasonable expectations of privacy and persistent geolocational tracking. Wilkins asked, for example, “What is the government’s interest in obtaining RID data from a drone operating entirely within one’s property?”
Ross responded that drones often go beyond the barriers of one’s private property and that other drones can also enter one’s property and cause safety and security risks. In reply, Walker quipped, “It seems a little bit extreme to say you can regulate what someone does with a drone on their own private property because what they do on their own private property might interfere with what a trespasser does.”
The next series of questions involved whether or not the FAA is currently tracking drones and triangulating data internally, which Ross conceded is possible.
Ross noted there was no evidence the FAA was actually monitoring drones and a line was drawn, in Karo, between installing a device and tracking someone with it. The judges seemed to indicate that was a distinction without a difference here because the installation of RID technology on a drone allows for persistent tracking, and the FAA created the RID authority so that the government could exercise it.
The argument culminated in this enlightening series:
Walker: I am going to ask you a long question and it will end with “What do you want me to do?” Assume I think this rule does not violate the rule in 99.9 percent of its applications because although Carpenter and Jones carved out some still narrow exceptions to this general rule, I think there is a general rule that what someone does in public is not private, and most of the time when people fly drones, it’s in public. Now even say above your own treeline counts as in public. But there may be times where someone might not even leave the walls of their own house and they are flying a drone on their own property below the treeline and that drone and that person cannot be seen by anyone unless someone were to trespass on their property. In that instance, I think your rule is unconstitutional. What do you want me to write in an opinion if I get to write this?
Ross: My understanding is that the rulemaking would have to be done again.
Rupprecht then had three minutes to reattack. He noted that the FAA failed to mention the special needs doctrine in the final rule, which Ross had pitched to salvage the rule, in the rule itself. Quoting the Supreme Court’s majority opinion in Motor Vehicle Manufacturers Association v. State Farm Insurance, he noted, “Courts may not accept appellant counsel’s post hoc rationalizations for agency action.” He continued, “There was a whole lot said trying to rescue this doomed ship, this Fourth Amendment case, but … they had the opportunity and they did not bite at that apple.”
A ruling is likely to come sometime in early 2022. Which party will come out on top remains to be seen.
Still, it speaks volumes that the judges pushed the government on the Fourth Amendment for three-quarters of the time allotted for both parties’ arguments. Perhaps even more telling, the government candidly conceded the FAA’s rule falls short on Fourth Amendment accounts in the limited circumstances Walker identified.
At the same time, the panel seemed to be inclined to agree that in 99.9 percent of the cases, the application of remote ID would pass muster. The judges even threw the government a lifeline, suggesting that possible RID carve-outs could be granted for ranchers and backyard flyers. But the government did not back away from its all-or-nothing approach.
Interestingly, the FAA also did not take the position in its briefs or arguments that it did within the actual RID rule: that it deferred to law enforcement and security agencies to abide by the Constitution. Presumably, that would include obtaining a warrant when required. As the case progressed, the agency simply went all-in on a “no warrant needed here” posture.
Should RDQ prevail, the rule would be vacated. The effect of an order would be a redo of the RID rule. The practical impacts of this would be profound. Law enforcement and national security agencies will have to continue waiting for a valid universal drone identification schema to detect and, as permitted, mitigate drone threats. The commercial drone industry will also take a hit because the FAA incorporated RID as a predicate technological requirement of drone operations over people and a future low-altitude privatized unmanned traffic management system.
It will also be interesting to see how the court will address the “avigation” issue. In its questioning, the panel seemed to presume that individuals had property and privacy rights below navigable airspace (500 feet above ground level) or, as they said, “below the treeline,” over one’s privately owned property, in certain circumstances. This could have ramifications for government officials desiring to control drones in such areas, as well as for the collection of evidence in such areas with their own drones. As for the commercial drone sector, companies like Amazon that want to overfly someone’s backyard for their drone delivery service might just have to pay for an avigation easement.
Regardless of the outcome in this case, balancing citizens’ rights with security, privacy and safety when it comes to drones will remain a tough challenge for years to come.