A95H0012-Controlled Flight into Terrain - Campbell River, B.C. - 27 September 1995

Safety Action Taken
(as presented in the TSB Report)

Seating and Restraint System

Subsequent to this accident, the TSB issued a safety advisory to Transport Canada (TC) identifying a concern that seating and restraint systems of some aging aircraft do not provide adequate protection to passengers in the event of a crash or forced landing. Aircraft systems are being modernized to extend their useful lives for commercial passenger-carrying operations, but these upgrades seldom include the improved passenger safety provisions consistent with contemporary standards. Thus, the TSB suggested that TC take a more systems-oriented approach in approving such life-extension programs.

Engine Condition Trend Monitoring

During the investigation, it was established that the ECTM program, which formed part of the approved maintenance program for the turbine engine installation on C-FEBX, had not been used as per Transport Canada's approval. It was also determined that some of the TC airworthiness inspectors responsible for the Western Straits Air maintenance system were not trained in trend monitoring programs. The TSB subsequently advised TC of this issue and suggested that TC consider adding ECTM to the airworthiness inspectors' training curriculum.

Safety Action Required
(as presented in the TSB Report)

Air Regulations and Air Navigation Orders established under the Aeronautics Act, such as those governing VFR and SVFR flights, prescribe operating limits. Such limits are designed to provide operational flexibility, while ensuring minimum acceptable safety margins. Such regulations are influential elements in determining industry operational practices and in establishing the level of safety of the transportation system.

The Campbell River accident raises questions regarding the feasibility of VFR and SVFR flights in marginal weather conditions, considering pilots' limited capability to recognize deteriorating visibility, the adequacy of the margin of safety afforded by VFR and SVFR regulations, and the level of operators' awareness of the risks associated with commercial operations in marginal weather conditions.

Visual Flight - Margin of Safety

In its 1990 Report of a Safety Study on VFR Flight into Adverse Weather (90-SP002), the Board made several recommendations to the Department of Transport concerning visual flight rules. In recognition of the problems of safely maintaining visual references in mountainous terrain, the Department of Transport subsequently increased VFR visibility minima in designated mountainous areas from one mile to two miles.

At the time of this occurrence, the aircraft was operating under VFR in uncontrolled airspace within a designated mountainous area. To be in compliance with VFR, the aircraft was required to be flown with reference to the ground or water and remain clear of cloud; the pilot was required to stay within sight of the surface of the earth at all times, and was required to ensure a minimum flight visibility of two miles. Under SVFR in the Campbell River control zone, the same restrictions would apply except that the minimum flight visibility would be one mile. There are no minimum ceiling requirements for either SVFR flight or VFR flight in uncontrolled airspace below 1,000 feet agl.

The accident record continues to show that VFR CFIT accidents occur primarily when operations are being conducted in marginal weather and/or dark night conditions. Interviews with flight crew and operators obtained during this investigation indicate that it is common practice in the industry to continue flight operations at the minimum visibility requirements for VFR and SVFR. When operating in conditions that include low ceilings and visibilities, pilots often pick their way through the weather while trying to stay visual. Flight in marginal weather presents a high risk of inadvertent entry into conditions where visual reference is insufficient for the maintenance of aircraft control, terrain and traffic avoidance, and accurate navigation. There are substantial grounds for questioning a pilot's ability to safely complete flights under such conditions.

Visual Depth Perception. In perceiving both depth and distance, humans interpret several visual cues in such a way as to generate a three-dimensional image in the visual cortex of the brain:

  1. Linear Perspective The distances between distant images appear to be less than those separating near images. For instance, railway tracks appear to converge in the distance. Knowing that the tracks remain a fixed distance apart, the convergence is interpreted as a distance cue.

  2. Clearness In general, the more distant an object, the less clearly it is seen. Further, a mountain on a hazy day appears more distant than it would on a clear day.

  3. Interposition When an object partially obstructs the view of another object, the first object appears nearer.

  4. Shadows Humans are used to perceiving objects with light sources situated above them; this information is used to give objects a spatial orientation.

  5. Gradients of Texture Generally, the texture of a scene appears finer and there is less detail as distance increases; conversely, foreground appears coarser and there is more perceptible detail.

  6. Movement When the head moves, objects move in relation to oneself and to each other. Objects beyond the eyes' visual fixation point move in the same direction as the head. Objects nearer than the point of visual fixation appear to move in the direction opposite to the movement of the head. The amount of movement is less for distant objects than for near objects.

Assessing Distance in Marginal VFR. When visibility is poor, the cues to perceive distances of objects are diminished. Without these cues, consistent, accurate judgement of distance is improbable, even in a relative sense.

Humans tend to be poor judges of distance in absolute terms; they can best judge distance in relation to some fixed marker. Thus, trained weather observers use known distances from ground features to establish ground visibility. In the mountainous area of Vancouver Island, distance cues or markers are much less likely to be available. Reliably judging one mile of visibility from a moving aircraft is arguably a task beyond human capability.

Another factor that could detract from a pilot's ability to judge whether the one-mile visibility requirement is being met relates to the angle of flight visibility being considered. The primary requirement of visual flight is that the aircraft shall be flown with reference to the ground or water. When an obscuring phenomenon like fog is present, the reduction in visibility at low altitude can be much less looking downward than looking forward. The survivors of the Campbell River crash reported that they were able to see the ground in the Campbell River area; however, obstacle avoidance requires forward visibility, which was not available. A reasonably clear view of the ground in marginal conditions could lead a pilot to believe that the one-mile forward flight visibility requirement was being met. Clear downward visibility would likely be an influential cue to the pilot, even though not necessarily a reliable or accurate cue.

Aircraft flight control and navigation may be conducted exclusively by visual reference to cues outside the cockpit, by reference to aircraft instrumentation, or by varying combinations of external and internal references. In this accident, the pilot was apparently attempting to avoid terrain and navigate by using a combination of external references and aircraft instrumentation.

In Canada, air-taxi operations are often conducted fully or in part under visual flight rules. Over an eleven-year period (01 January 1984 to 31 December 1994), there were 70 accidents involving commercially operated aircraft not conducting low-level special operations, where the aircraft were flown into terrain, water, or obstacles, under control, while the crew had no awareness of the impending disaster. In over half of the occurrences, the crew was attempting to see the ground in order to fly visually, although the conditions apparently precluded visual flight. These 70 CFIT accidents involved pilots with the full range of experience, indicating that experience does not appear to be a factor in reliably coping with conditions of marginal visibility. Several recent commercial accidents (A95P0268, A95C0026, A95Q0104) in which VFR flights were attempted in flight conditions which did not allow the pilots to visually navigate and/or avoid collision with terrain illustrate that the same issues continue to be factors in this type of occurrence.

The Board understands that the present VFR and SVFR requirements are the result of years of evolution in committee proceedings; however, these committees seldom include representation from the scientific community, and therefore do not take into account the research available on such aspects as the human visual system and normal human information processing capabilities. Safe operations in VFR and SVFR conditions depend almost entirely on the pilot's ability to assess flight visibility and immediately recognize any deterioration. When flying in the minimum weather conditions allowed by VFR and SVFR, recognition of deteriorating visibility can be virtually impossible, particularly when combined with other factors such as high workload, variable weather conditions, poor light conditions, or limited outside visual cues. Therefore, the Board recommends that:

The Department of Transport sponsor research to establish on a scientific basis the ability of pilots to assess distances, make appropriate decisions, and control aircraft without reference to aircraft instruments in the marginal visibility conditions of VFR and SVFR minima. (A96-09)

Transport Canada's Response:

Transport Canada (TC) will sponsor research to establish on a scientific basis, the ability of pilots to assess distances, make appropriate decisions, and control aircraft without reference to aircraft instruments in marginal visibility conditions of VFR and SVFR minima. In addition, human factors expertise will be utilized to ensure that the scientific data is linked to the unique stresses found in a pilot’s working environment that may impact on a pilot’s ability to make proper decisions and to exercise sound judgment.

Date modified: