Order In the Skyways: A Comment on the Regulation of Drones

The increasing popularity of drones is attracting the attention of the regulatory process as municipalities, such as Calgary, attempt to control the use of drones in public areas through the bylaw process (see section 24(c) of the City of Calgary Parks and Pathways Bylaw 20M2003). In fact, the issue has become so pressing that the federal government recently announced immediate action through the Interim Order Respecting the Use of Model Aircraft  by amending the Aeronautics Act RSC 1985, c. A-2 to more specifically address the “significant risk” the operation of drones have “to aviation safety or the safety of the public.” However, regulation in this nascent area of recreation has not been without difficulties. The extent to which the regulatory regime can effectively and fairly maintain order in the skyways may appear a simple task but as with any statutory process, “the proof is in the pudding” or as in the recent summary conviction appeal against conviction in R v Shah, 2017 ABQB 144 the “proof,” involving the appropriate application of the standard of proof, was lacking.

The Appellant in this case, ably represented both at trial and on appeal by our very own Student Legal Assistance, was flying a recreational remote controlled drone during the evening hours of January 16, 2016 when he was charged under section 602.45 of the Canadian Aviation Regulations SOR 96/433 enacted under the Aeronautics Act. The section states that: “No person shall fly a model aircraft or a kite or launch a model rocket or a rocket of a type used in a fireworks display into cloud or in a manner that is or is likely to be hazardous to aviation safety.”

R v Shah is an appeal of a summary conviction entered at trial in Provincial Court.  Three grounds of appeal were advanced before Madam Justice J. Antonio (at para 2). The first ground concerns the trial judge’s application of the standard of proof. The second ground involves the proof of the required actus reus elements of the offence, namely the requirement the model be flown “in a manner...likely to be hazardous to aviation safety.” The third ground raises the issue of reasonable apprehension of bias on the basis that the trial judge “pre-determined” the case. The first ground was successful while the other two were dismissed. Justice Antonio has ordered a new trial.

The facts are brief but provide insight into the issues raised on appeal. While on road patrol, two officers observed “some blinking lights in the dark sky” above a park located “just south” of the main runway of the Calgary International Airport (at paras 4 and 6). The officer noticed the lights were above the trees, which the officer “estimated” as 80 feet tall (at para 4). When the officers arrived on scene located at a nearby park, the Appellant was packing away his recreational drone into his car. While the officer was conversing with the Appellant, passenger planes were “coming in for landing with wheels down” at an “estimated” altitude of 200 to 250 feet (at para 6). The officer, did not however, notice any planes in the air when the drone was in use. According to an officer who later tested the Appellant’s drone, the object could fly up to an “estimated” height of 200 feet (at para 7).

The officers were not qualified to give expert evidence and no other evidence was called by the Crown in support of the prosecution. Although initially the Crown attempted to rely on a regulatory requirement imposing a 9 km no-fly zone for drones near an airport, an adjournment revealed that in fact there was no such regulation in place at the time of the incident. Ultimately, the trial judge convicted the appellant of the offence in a brief oral judgment.

Turning to the first ground, the error which resulted in an order for a new trial, Justice Antonio found that in convicting the appellant the trial judge failed to correctly resolve the required standard of proof. It was argued, and referenced by the trial judge, that the matter, as a regulatory offence, was one of strict liability. A strict liability offence, as described by Justice Dickson at page 1326 of the seminal case on regulatory or public welfare offences R v Sault Ste Marie [1978] 2 SCR 1299, is an offence for which the Crown need not prove mens rea as the proof of the conduct is prima facie proof of the offence at which point the burden then shifts onto the defendant to establish a due diligence or mistake of fact defence. Strict liability offences are, as Glanville Williams explained in Criminal Law (2d ed.): The General Part at page 262, a form of negligence with the focus on whether the defendant, as a reasonable person, took all due care in performing the legitimate yet potentially harmful activity.

This form of liability was reviewed under section 7 of the Charter in R v Wholesale Travel Group Inc, [1991] 3 SCR 154 where the Supreme Court unanimously concluded that the Charter requirements are “met in the regulatory context by the imposition of liability based on a negligence standard” (Justice Cory at p 241). However, as proof of the actus reus elements of the offence is key to conviction and considering the Charter values reflected in the presumption of innocence, the Crown, is obliged to prove the actus reus components of a regulatory offence beyond a reasonable doubt (at p 248). This basic proof principle was not clearly recognized by the trial judge in the Shah case when he stated in his reasons for conviction that the Crown “met the burden of establishing” the actions of the Appellant to his “satisfaction, and I’m not certain whether it’s beyond a reasonable doubt” (at para 13). It was this clear error, as found by Justice Antonio, which resulted in the quashing of the Appellant’s conviction and the ordering of a new trial. Hopefully, this decision will highlight this key requirement in future such prosecutions. Certainly, it is the obligation of the Crown to articulate this standard in their submissions.

The second ground of appeal was dismissed by Justice Antonio but still deserves attention. Essentially, the Appellant’s argument was that the Crown was required to prove the drone was operated by the Appellant “in a manner...likely to be hazardous to aviation safety” by way of expert evidence, and that by not doing so the Crown failed to provide any evidence upon which the trial judge could reasonably convict. The Appellant further alternatively argued that the trial judge erred in his interpretation of the term “likely.” Justice Antonio collapsed these two issues and entered into a statutory interpretation exercise, focusing on the word “likely” as it modifies the term “hazardous.”

Through this approach, Justice Antonio found that the required elements of the offence are directed to the likelihood of an aviation hazard, implying “a risk of a risk” and therefore casts “a broad actus reus net” (at paras 21 and 22). On that basis, considering the wide range of possible hazards, Justice Antonio found expert evidence on that issue may or may not be required depending on the facts of the case. As an example, she referred to the R v Khorfan, 2011 ABPC 84 (CanLII) decision wherein Judge Fraser considered expert testimony on the use of a halogen spotlight a “hazard” under the regulations. In Justice Antonio’s view, it would “fall to the next trial judge to determine whether the evidence before him or her” was sufficient to prove the conduct fulfilled the offence requirements of “likely hazardous” (at para 25).

Although Justice Antonio’s reasoning on this issue is attractive and provides for a robust reading of the section consistent with the objective of the regulatory scheme as aviation safety, the salient issue here is not how “likely hazardous” may generally be proven but whether in this specific fact situation there was evidence upon which a properly instructed trier of fact could reasonably convict. In this case, the only evidence before the court was from two officers who had no particular knowledge or expertise in estimating height, never mind any ability to “estimate” the height of an airplane at night in the backdrop of a darkened sky. As cautioned in Graat v The Queen, [1982] 2 SCR 819 (CanLII) police officers testifying in this manner are giving a compendium of observations, which are merely factual observations similar to any other witness’s factual observations. Their evidence is not enhanced by their position as police officers. In other words, these officers were in no special position to make the height estimates. Officers may have some expertise in assessing speed and vertical distance due to their daily role as police officers who enforce traffic regulations, however. they have no particular expertise in assessing altitude. This is not a road hazard but an aviation safety hazard.

In fact, the federal aviation regulations (section 602.14 of the Canadian Aviation Regulations) specify aircraft height maximums when flying over a built-up area. Although the regulations also indicate that those maximums do not apply when the aircraft is landing, there is absolutely no evidence in the Shah case at what point any of those observable aircrafts were in their landing and therefore not subject to the regulated height restrictions. The issue here is not just if the conduct was a “likely hazard” but is this conduct, as observed by the officers on the evening in question, “likely hazardous to aviation safety.” To prove the risk of the conduct to aviation safety beyond a reasonable doubt more cogent evidence was needed to connect the inferences than simply a lay person’s belief that this drone “might” reach a certain “estimated” height and “might” therefore be a hazard to an airborne aircraft which “might” be at another height. Proof beyond a reasonable doubt cannot rest, without more, on such imprecise opinion evidence dependent on an imprecise chain of reasoning. In Professor Wigmore’s words (7 J. Wigmore, Evidence, 1917 at 1-2.), a witness must be a “knower, not a guesser.” This evidential requirement is even more important considering the new changes announced to the aviation regulations which provide for precise statutory limits within which a drone cannot be flown. The new legislation prohibits the flying of recreational drones “at an altitude greater than 300 feet AGL” or “within 9 km of the centre of the aerodrome” circumscribing what would constitute a likely hazardous situation. It is interesting to note that the height in the new regulations differs from the height “estimates” provided by the police officers.

Clearly, there needs to be reliable and probative evidence on the record to make these findings be it expert evidence or not. In the Khorfan decision, for instance, the prosecution called non-expert evidence from an air traffic controller who was on duty the evening of the offence and gave crucial evidence on “aircraft taking off and their height and angle” (at para 30) at that time. This kind of evidence could provide the needed connection between the drone flight altitude and the “likely” aviation safety hazard. This argument ties into the first ground of appeal where concerns were rightly made with the trial judge’s application of the appropriate standard. There was simply no evidence upon which to convict beyond a reasonable doubt.

The final ground of appeal concerns the trial judge’s role as an impartial arbiter of the case, a role which is at the heart of our adversarial system. The test for unreasonable bias for good reason requires a high standard and Justice Antonio correctly articulates the difficulties in advancing such a ground, particularly in provincial court where justice must be done not only fairly but also efficiently. On a certain view, however, this ground is also linked with the first and second grounds of appeal in that the failure to properly apply the required standard of proof resulted in findings which may appear cursory and “pre-determined.”  

In the end, this prosecution was unsatisfactory. Considering the rise in regulation of our day to day activities, this case should be viewed as a caution that even in the realm of public safety, convictions must be based on sound principles and evidence.