Only five percent of general aviation accidents during 2010 occurred in instrument meteorological conditions (IMC). But these accidents comprised 18 percent of all fatal crashes that year, and almost two-thirds of them proved fatal. According to the AOPA Air Safety Institute’s 2011 Nall Report, this is a familiar pattern.
The charts on the opposite page summarize some of the data in the Report. For our purposes, it’s startling to realize 54.3 percent of non-commercial fixed-wing accidents in 2010 involved aircraft with at least one instrument-rated pilot aboard. How does that happen? Doesn’t the instrument rating teach us how to handle weather flying? Isn’t the instrument rating supposed to make us safer? Three examples of weather-related problems encountered by instrument-rated pilots may be informative.
“It looked dark,” but other pilots were reporting the route was VMC, the pilot of a Beechcraft A36 Bonanza later said. He was a regular volunteer, flying medical doctors to remote locations in Mexico as part of a charitable flying organization. He had extensive experience in his airplane and in instrument operations, and had a personal requirement of completing a challenging instrument proficiency check every six months.
On one of his trips as a volunteer, a tropical storm threatened. He and the other pilots who had flown physicians down for a weekend clinic in single-engine airplanes were anxiously watching its approach. As departure day dawned, there were light showers in the area and the pilots were “a bit concerned;” it was “hard to define the bases” of the clouds but the tops of mountains in the area were still visible.
One pilot who had flown to this location for over 10 years said he had never seen rain there, so none of the pilots had relevant experience with what the combination of humid air and rugged terrain would mean. The two closest weather reporting points were both 50 miles to the southwest, toward the storm. Both reported a 6000-foot scattered layer and a 12,000-to-14,000-foot overcast in good visibility. A call to an airfield about 100 miles north, in the direction home, revealed clear skies and unlimited visibility. The indefinite conditions were apparently only localized.
Airspace around the departure airport and up the long valley northbound was uncontrolled. Similar to uncontrolled airspace in the U.S., instrument-rated and current pilots in IFR-certified airplanes may fly IFR in uncontrolled airspace, but there is no clearance, no ATC and no traffic separation. Terrain avoidance is entirely the pilot’s responsibility. None of the pilots contemplating departure, all of whom were instrument-rated and current, considered the uncontrolled-IFR-in-mountains choice to be a viable option. The charitable organization did not specify go/no-go criteria, so it was left to each pilot to make a decision. Faced with the approaching storm, one by one they decided to load up their doctors and go.
Three airplanes departed before the Bonanza pilot. One, a Cessna 182 flown by a Naval Aviator, radioed back that he quickly reached clear skies with good visibility. He did not mention, however, that instead of following the valley northward as was usually done, he had skipped over some low terrain to the west and then turned up the coastline on the upwind side of the route. Expecting to find clear conditions between the peaks flying up the valley, the Bonanza pilot took off with his passengers.
Instead, the Bonanza quickly entered “heavy rain” that severely limited visibility. Knowing there was a ridgeline topping 2500 feet msl just west of his course, and a 6000-foot peak somewhere to his right, the pilot leveled at 2900 feet as his visual contact with the ground diminished in the heavy rain. The pilot “became concerned he had not confirmed the height of the terrain” on his route—he knew he had “high stuff” to the left and “higher stuff” to the right, but he did not know precisely where.
Suddenly the airplane plunged into cloud. The pilot, now “very concerned about height,” remembers the sensation of seeing a mountain straight ahead—he later learned the illusion likely was the result of his fear. Immediately upon perceiving the obstacle just ahead, however, the pilot pulled back on the controls to gain altitude. He does not remember adding power.
The passengers later reported the Bonanza’s stall warning horn sounded for “a long time” before the pilot “pushed hard,” resulting in “negative Gs.” As suddenly as the airplane entered clouds it was out of them again, descending into the valley. The rain slackened and the pilot could see a light area ahead; he aimed for it and soon broke out into clear air. The flight continued uneventfully.
The Bonanza pilot makes no excuses. He accepts responsibility for his actions and won’t let it happen again. His overall reaction is that he “can’t believe it happened. How does this happen,” he asks, “to a qualified and [instrument-] rated pilot?”
The professional pilot of a Cessna 414A departed Texas for an airport in the Colorado mountains. It was a flight he’d made often with the airplane’s owner as a passenger, as was the case on this trip as well. An approaching snowstorm pressed the afternoon departure; the freezing level was near the surface with cloud tops to 27,000 feet. Heavy snow was forecast for the vicinity of the destination shortly after the ETA.
As the flight neared its destination, the pilot told ATC he’d like a visual approach if the ceiling was above 1000 feet. Center replied that another aircraft had just completed the ILS and broke out at 800 feet agl, right at minimums. The pilot then requested and was cleared for the ILS approach.
A review of onboard GPS data indicated the twin Cessna flew through the final approach course several times during the procedure, and was consistently below the glideslope. The airplane descended below decision height and then drifted to the right. The GPS recorded the pilot’s attempt to perform a missed approach: a rapid decrease in groundspeed and then descent to the ground “consistent with an aerodynamic stall.”
The airplane owner, one of four aboard who survived the crash with serious injuries (the pilot and a fifth passenger died), recalls “two left-turning circles” before impact. Examination revealed no pre-impact anomalies. The pilot had not activated ice-protection equipment. Although an autopsy indicated the 75-year-old pilot suffered from a severe heart-related condition, pilot medical issues were not considered to be a contributing cause. The NTSB concluded “it is likely that the airframe collected ice during the descent and approach, which…led to an aerodynamic stall.”
The pilot of a Piper Cherokee Six told ATC he was diverting around a growing area of thunderstorms. Flying at 8000 feet, the pilot reported he was in “bad” weather (ground radar reported heavy rain showers in the area), and that he was “trying to get out of it.” A witness on the ground heard a sound like “an explosion” just before light drizzle turned to heavy rain “pouring down.” The airplane, minus its left wing, vertical stabilizer, rudder and right wing tip fuel tank, impacted about 450 feet from the witness’s house. All five aboard the aircraft perished.
The pilot owned a portable GPS with a current subscription to a satellite-based Nexrad uplink service. The GPS unit was located in the wreckage. The pilot had been receiving radar weather information from ATC but disregarded the controller’s suggestions in favor of his onboard Nexrad, even when the controller advised the pilot’s plan was taking him straight into severe precipitation echoes.
According to the GPS device’s owner’s manual, “Nexrad weather data should be used for long-range planning purposes only,” and warns not to “penetrate hazardous weather” using Nexrad uplinks as a guide because “Nexrad data is not real-time.” In part as a result of this investigation, the NTSB issued a Safety Alert about “misleading” Nexrad data because of the lag between observation, transmission to the cockpit and update cycles. The NTSB found the crash resulted from the pilot’s “inadvertent encounter with severe weather,” in part due to “the pilot’s reliance on outdated weather information” from the Nexrad uplink.
No way out
These two fatal crashes and the inadvertent encounter with IMC that resulted in an aerodynamic stall—which could easily have been fatal had the airplane not broken out of the clouds with sufficient altitude to recover—all involved experienced, instrument-rated pilots flying well-equipped IFR airplanes. In all three cases, the pilot “painted himself into a corner,” i.e., he apparently made a conscious decision that put the airplane in a place where there was little to no opportunity for escape. In other words, the pilots violated one of the tenets of flying and failed to leave themselves an “out.” How can that happen? Don’t we all chant “give yourself an out” at every training event and flight review? And more importantly, what are we forgetting that allows the experienced among us fall into this closing trap?
One key, I believe, is the inherent difference between the way we approach training events and our day-to-day flying. As I co-wrote with Dr. Lorne Sheren in the November 2011 issue of AOPA Pilot, we are pessimists when we train and optimists when we fly. In other words, during instruction, we’re spring-loaded to look for anomalies and indications of problems. We expect (simulated) engine failures and partial panel operations and alternator outages. Go-arounds and missed approaches are expected and executed; delays and diversions are part of the plan.
Take us out of the training environment, however, and pilots become wildly optimistic. We’re on a mission to make it to a destination on schedule, and we tend to rationalize away anomalies. Weather is something to master, not avoid. To the point of our three examples, we expect things will work out. Not only do we fail to plan for contingencies, we fail to recognize the need for a plan.
The Bonanza incident and the Twin Cessna and Piper crashes remind us that we train for weather delays and diversions for a reason—we need to be ready to execute our options in the real world, not just during instruction. More so, we should be proactively evaluating our “outs” just like we would in a training scenario, and exercising an “out” before we find ourselves with only one or no options remaining.
Briefing your out
I evaluate weather in terms of the four major aviation weather hazards: thunderstorms, turbulence, low clouds and reduced visibility, and ice. I look at the likelihood of each hazard, and pick a route and altitude to minimize exposure to the risk…or avoid it completely, in the case of thunderstorms and ice.
If my flight takes me near the boundaries of a threat, which includes an approach if the hazard is low clouds and/or reduced visibility near the surface, I’ll brief the action I’ll take if I begin to encounter the threat. “If I begin to pick up ice I’ll climb to above the clouds. If that doesn’t work, I’ll immediately turn back to ice-free air. If I hit moderate or greater turbulence, I’ll immediately change altitude.” “If ATIS reports the visibility below minimums I’ll go ahead and proceed to my alternate,” etc. Before I ever take off, I’m taking a pessimistic—let’s call it “healthy”—look at my options. Considering them beforehand, I’m more likely to exercise my “out” before running out of options.
In flight, as I near areas of adverse weather I’m always asking myself: “Which way will I go if it starts to get too bad?” For example, when flying in the Midwest, it’s very common in the spring and summer to find myself deviating around cumulus buildups. My goal is to remain visual between the clouds; I’m constantly telling myself, “If I start to hit moderate turbulence I’ll turn left,” etc. In cooler weather, I pick routes and altitudes to avoid potential icing conditions. But sometimes the temperature drops when you enter a cloud, and icing can threaten when you did not anticipate it. So as the temperature nears freezing, I’ll work to stay out of the clouds, consistently reviewing my escape plan at the first sign of ice accumulation.
It’s merely a matter of flying like you train—including a sharp outlook for trouble, and a quick reaction with your pre-planned response when conditions require a change in plans.
One last commonality between the three cited events: each pilot was trying to complete the flight before conditions got worse. Very often in weather-related accidents, the pilot accepts what turns out to be unacceptable weather out of fear conditions will worsen. Instead of speeding up to try to fly ahead of the bad weather, how about slowing down to let the hazards go by?
Far too often, weather-related fatalities occur when the pilot appears to have no option left. The reality is they allowed themselves to get into that situation by forgetting what they learned in training.