March 1, 2001
By Alton K. Marsh
Lack of aeronautical information was killing too many pilots in 1930 and 1931, prompting Elrey B. Jeppesen to start his famous black book. He filled it with intelligence information on every pilot's enemies — terrain, smokestacks, water towers, and other obstacles. He sold copies for $10, and the sales became the start of Jeppesen, now a Boeing company.
Today, the enemies remain the same, and although there is no lack of information about them, pilots are still dying. The reasons include failure to follow approach procedures or to recognize personal and aircraft limitations, and lack of respect for the weather. That's the story told in 21 accident reports gathered for this article by the AOPA Air Safety Foundation (ASF). The accidents involved 53 people: Twenty-one of them died, while another 21 were injured — some seriously. The rest stepped uninjured from damaged aircraft. The following 10 "ounces of prevention" might have prevented those accidents.
A common thread running through all 21 accidents is the willingness to descend below minimum altitudes, either on the first approach or after two or more missed approaches. Decision making was often based more on convenience than safety. Pilots had to get to their workplace, or had to get the passengers to their original destination, or had to at least land near the original destination without regard to aircraft and pilot capabilities — weather be damned.
A Piper Comanche was approaching Virginia's Richmond International Airport, in 1992 as the weather deteriorated. During the first attempt at an ILS approach the pilot told controllers that his ILS receiver was not working, and he requested an airport surveillance radar approach. That was granted but resulted in a missed approach. Seconds after the missed approach began, the pilot transmitted, "Just a little bit lower. We just saw the runway lights." He then flew a quarter-mile past the runway, hit trees, and crashed onto a road. Both aboard were killed.
Failure to execute a missed approach was cited by the NTSB in 15 of the 21 cases studied as a probable cause of the accident. In most of the accidents, it wasn't a matter of the pilot being unprepared for the procedure. Rather, it was the often fatal decision by the pilot to force the approach into a landing. Nearly all of the 21 accidents involved weather that was at or below minimums.
A better method might be to rig the approach in your favor before leaving home. That is, be sure that the forecast weather is better than you need — at least double the minimums — to complete the approach to a landing. It is understandable that pilots want to avoid the missed approach, as it requires a change in aircraft configuration and attitude while navigating the procedure itself. To avoid it, be sure that you won't need it.
How low will you go? That depends on your current level of skill. Set personal weather minimums at levels that you are confident you can handle, even with an equipment failure.
Challenging the weather to a duel is the second most common cause of approach accidents, according to our sample of 21 accident reports. The pilots knew the weather was below minimums before attempting the approach. There's no problem with making the approach in order to take a look, but why bother? You're not going to see anything.
A Cessna 182 pilot had flown passengers along the route to Edgartown, Massachusetts, at least 10 times, and was getting his usual $200 to make the trip. The weather that September night in 1992 included an indefinite ceiling of 100 feet and one-half mile visibility in fog. The pilot flew an unstabilized approach that included errors of airspeed, rate of descent, and position with respect to both the localizer and glideslope. The aircraft descended below minimums and crashed a half-mile short of the runway, killing him and his two passengers.
In another case, an experienced pilot of a corporate jet found the weather to be below minimums at Concord, North Carolina. He then had a choice of landing 53 miles to the north at an airport that had a ceiling of 900 feet and six miles' visibility or at Charlotte, North Carolina, which was closer to the customer's destination. The weather at Charlotte was an indefinite ceiling of zero and a visibility of one-quarter mile: He chose Charlotte. The aircraft continued below decision height without any visual reference to the runway environment and crashed, killing the pilot and seriously injuring the passenger.
Some accidents included pilots who set their radios incorrectly and never caught the error. One pilot died after attempting a localizer/DME approach with the DME switch set in the wrong position.
Use a checklist for the things you want to accomplish when setting up the approach. It weighs less than an ounce, but provides tons of accident prevention. Prior to arrival, pull out your homemade checklist. It can include checking the ATIS at the arrival airport, comparing the actual weather with the minimums on the terminal procedure chart, and setting radios. You should also read the name of the terminal procedure chart to make sure it is the same as the approach that you were assigned.
Another pilot in 1997 was familiar with the business jet that he flew, but set the DME switch in the wrong position for an ILS/DME approach to Salt Lake City. The aircraft was observed to be on the glideslope for 28 seconds before stalling and crashing 1.3 miles short of the runway. The jet had been too high during most of the approach. One person died and three were injured.
Airline pilots report that they are rusty at making approaches after only a two-week vacation, while most general aviation pilots insist they are still sharp after months have passed. Don't get mental instrument skills confused with motor flying skills. Flying skills are retained much longer than instrument skills. Sure, you remember being good at approaches. But if that was in December and this is March, you're in for a surprise. A cheat sheet brings a little confidence to the process before you begin the approach.
Many of those involved in approach accidents in our sample got in trouble because they changed the plan, whether it was the published approach procedure or their own decision making.
A Cessna 206 pilot approaching Long Beach, California, in October 1995 initiated a missed approach. A fog bank was slowly overtaking the airport. During the missed approach he saw another runway that was still clear of fog and was cleared by the tower to land on that runway. While maneuvering in visual conditions, he stalled the aircraft at low altitude and hit terrain. He was not injured. The NTSB cited as a probable cause his failure to continue with the missed approach.
While shooting a VOR approach to minimums at Delano, California, in December 1995, a pilot became disoriented. A vector back on course was provided, and the approach was completed into VFR conditions. While circling to land, the Beech V35 Bonanza entered the clouds and later was seen to emerge in a steep dive that ended in a crash. After purchasing the Bonanza, the pilot had completed a 15-hour VFR checkout, but the instructor suggested that the pilot get more experience and then get an instrument checkout. Instead, the pilot took an instrument proficiency check in a simpler Piper Archer. The NTSB listed the pilot's lack of instrument flight experience in high-performance airplanes as one of the probable causes. Four people died.
A working autopilot capable of flying coupled approaches might have saved several of the pilots listed in our accident reports. An autopilot is essential when handling single-pilot emergencies such as emergency extension of the landing gear. Renter pilots often complain that checkouts rarely include the autopilot; the instructor may not even know how to operate it. Take the autopilot manual home and review it.
Some of the accidents involved pilots who simply couldn't fly instruments. A pilot approaching Kneeland, California, in 1995 became spatially disoriented. While inbound, the aircraft was seen to turn left while left of the localizer, and then right. The right turn continued into a spiral and resulted in a crash. A person who had flown with the pilot said that he sometimes turned the wrong way, and at times seemed to have difficulty in simply maintaining aircraft control.
Aside from keeping instrument skills sharp, buy an approach-capable autopilot and turn it on.
Fog was a factor in six of the 21 accident reports provided by ASF. Four of those resulted in serious injuries or death. Fog brings the lowest of all ceilings and the greatest unpredictability. Unless you have a Boeing 777 that can land automatically and brake to a stop by itself on the runway centerline, don't mess with fog.
A Beech King Air pilot tried his luck with fog at Olney, Illinois, in 1995 by descending below the minimum descent altitude while on a localizer approach. The pilot may have been lulled into thinking that the weather was better. Since controllers could not provide Olney weather, they provided the weather from Evansville, Indiana, 20 miles away. Evansville had 1,200 feet scattered and 4,500 overcast with five miles' visibility. Olney, however, had partial obscuration with a 100-foot overcast and visibility of three-quarters of a mile. The King Air continued to descend after passing the airport until it struck the ground, killing the two people aboard.
Another pilot and passenger survived their encounter with fog and drizzle at Lago Vista, Texas, in 1995. The Beech Bonanza completed a VOR approach, but the pilot did not see the runway until he was "almost on top" of it, at which point he commenced a tight spiraling approach. He continued the spiral below 300 feet with a high rate of descent. The unusual approach ended with the right wing hitting the runway.
Easier said than done, right? Now that we have highlighted pilot decision making as a cause of accidents, it's time to consider the controllers' role. Pilots who fly for a living and get to know one airplane very well may not be fazed by a controller's request for a slam-dunk approach. The rest of us may be too proud to admit to the controller and other pilots listening on the frequency that we are not prepared for an unstabilized, razzle-dazzle arrival.
An accident at Chino, California, in September 1995 fortunately resulted in only a minor injury. The weather was one-eighth of a mile with fog. The minimum published visibility for the approach was three-quarters of a mile. The controller gave the Swearingen SA-226T an intercept angle to the localizer that was greater than the maximum allowable and three miles closer to the runway than the actual intercept point should have been. The pilot ended up 650 feet above the glideslope, an altitude that was outside the glideslope parameters. The pilot tried to dive for the glideslope, developing in the process an excessive rate of descent that carried him through the glideslope and into the ground 1,000 feet short of the runway.
An approach should feel more like skiing the bunny slope than one marked by double black diamonds.
Just say no. There's a price for that, but mostly it's a social one. Sometimes it can be careerlimiting.
A pilot was asked by his boss to transport personnel in a twin-engine aircraft in icing conditions. The airplane was equipped for flight in known icing conditions. The pilot expressed reservations, but the boss insisted. So he went, and had to use alternate air when the engine air intakes iced over. After landing, he found three inches of ice in the engine cowlings. The deicing equipment worked, and he had no aerodynamic problems from ice during the trip, but he plans to refuse the trip next time, no matter what the consequences.
Don't forget that an IFR departure can result in an IFR approach back to the home airport. Don't count on escaping poor weather at your departure airport to better weather at the destination. I once refused to punch into a 600-foot overcast with two passengers in an airplane filled with new avionics. Not only were the avionics unfamiliar to me, but the avionics technician had said days earlier that he was suspicious of the aircraft's ILS receiver. I could imagine myself with an emergency — such as a door that pops open after takeoff — and fumbling with confusing new avionics to tune in an ILS frequency on an inoperative receiver. One of four fellow pilots who later ribbed me about my cowardice (you could land a 747 on that yellow stripe down my back) promised to take any future flights that made me uncomfortable. But I say, cowards, unite! Unless, of course, you're afraid to.
If you really have to go, you may already have forged that first link in an accident chain. Two cases of gotta-go-itis already have been mentioned. A third occurred at Fullerton, California, in November 1995 when an instrument instructor flew a localizer/DME approach but did not have a DME. He asked controllers to notify him when he was over the final approach fix, and after notification he began his descent. The Piper Saratoga crashed into an apartment building one mile short of the runway, killing the two on board and one person on the ground. The instructor was due for work near the airport 25 minutes after the accident time. The NTSB listed as one of the probable causes the instructor's self-induced pressure to get to work on time.
The ideal situation is to tell passengers you will go on one of three days, if you are flying aircraft that are limited to lower altitudes, and then pick the day with the best weather. You probably won't wait long for the best weather. NTSB accident investigators have noted that the weather for their investigations is nearly always clear — one day after a weather-related accident.
Many of the 10 ounces of prevention above may seem impractical or unrelated to your actual flight operations, such as when the forecasters badly miss a prediction that changes for the worse after you take off. But take them along on your next trip. After all, the whole package weighs only 10 ounces.
Links to additional information about making safe approaches may be found on AOPA Online ( www.aopa.org/pilot/links/2001/links0103.shtml). E-mail the author at firstname.lastname@example.org.
AOPA Pilot Senior Editor Alton Marsh has been a pilot since 1970 and has an airline transport pilot certificate and instrument and multiengine flight instructor certificates, aerobatic training, and a commercial seaplane certificate.
Safety and Education,
Aeronautical Decision Making,
Pilot Training and Certification,
The FAA encourages pilots to do a number of things in order to increase safety, but does not require them. Check out these three actions that are recommended.
Among the very first lessons a pilot learns is that a control yoke is not a steering wheel. Research underway in Europe could change that.
Your CFII usually follows up route-planning drilling with a review of appropriate regulations, and today’s selection is 14 CFR 91.185, "IFR Operations: Two-way radio communications failure."
VOLUNTEER AT AN AOPA FLY-IN NEAR YOU!
SHARE YOUR PASSION. VOLUNTEER AT AN AOPA FLY-IN. CLICK TO LEARN MORE >>>
VOLUNTEER LOCALLY AT AOPA FLY-IN! CLICK TO LEARN MORE >>>
BE A PART OF THE FLY-IN VOLUNTEER CREW! CLICK TO LEARN MORE >>>