If a pilot wants to establish a personal minimum for weather on approaches, it might be best to not leave that minimum altitude unless the runway itself is in clear view.
While this was being prepared there were two IFR accidents on ILS approaches, both at night. One airplane hit two miles short of the runway, the other hit "approximately 3,575 feet" short, according to the NTSB.
Accidents like this are far from uncommon on precision approaches. Why do pilots go below the electronic glideslope and crash? In both of these cases it happened before the pilot even reached the point where a decision had to be made. We can only imagine what would make a pilot go below the electronic glideslope well before reaching the point on the approach slope where the decision altitude would have been reached. Perhaps the indication was misread. More likely, something was seen that prompted the pilot to go lower for a better look. That is never a good idea.
The margins on an instrument approach are not great and they become ever less the closer we get to the appropriate minimum altitude. Pilots need to fly with that in mind.
Back to VFR. In the recent past there have been a number of serious accidents in pistons and turbine airplanes that have come during botched approaches and landings. When the event turns ugly, it is usually a product of the approach being too high and too fast, the pilot forcing the airplane onto the runway, the pilot deciding there is not enough runway left to stop, and the airplane crashing off the end of the runway because the go-around was started too late for any chance of success. All the margins were given away.
For a remedy to this we have but to look at an ILS approach. The decision height is 200 feet above the ground. On any approach on a clear day, that IFR minimum can be used as a defining point in determining if the margins are being maintained on the approach. At that time, if the airplane is on an electronic glideslope or the pilot is using a VASI and the indication is correct, then the altitude is correct. If the airspeed is correct then the margins are as good as they get. If the airplane is high or fast and the runway length is the least critical, nobody can know how much, if any, of the margins will remain if the approach is continued. That means that it should be abandoned.
When we look at takeoff and landing distances in the Pilot's Operating Handbook, the values reflected there are minimums. They might be accurate if you do everything perfectly, and if the engine is strong as new for takeoff or the brakes are as good as new for landing. Prudent pilots add margins. To keep it simple, I like to double the POH numbers for required runway length. Over the years, that has kept me from breaking a sweat on takeoffs and landings.
Certainly if you are looking at a takeoff and you calculate that the distance over a 50-foot obstacle shown in the POH is barely adequate, then you are flying with no margins and the chance of success is not good. The same thing is true of a landing. No margins, no good.
The one thing the accident record illustrates is that if on landing you put the wheels on the ground, you should let the wheels stay on the ground. There are that relative lot of serious accidents related to go-arounds started after a touchdown with a questionable amount of runway available for stopping. There are also a lot of accidents where airplanes run off the end of the runway but the main casualty in those is usually the pilot's ego. All accidents have the potential of being serious, and there's usually a point in the sequence of events where the pilot has a choice between a decision that results in a serious crash and one that leads to a fender-bender. Making the best choice depends on whether or not the pilot has given adequate advance thought to such situations and is locked and loaded to do the best thing.
There are a lot more choices on minimums, maximums and margins. The minimum level of pilot proficiency is defined in the FARs. So many takeoffs and landings and so many instrument approaches in a specified time define the FAA's version of current, but a pilot who is satisfied with being minimally current is really signing up for second-best. Going beyond the minimum requirements on anything means the margins have been increased.
Mac wrote at length about wind in his January column. And an important question has always been about how much surface wind is too much wind. The only guidance here is the maximum demonstrated crosswind that is not a limitation. The pilot of a Cessna 172 found his maximum wind on 11/14/08. According to the FAA, the wind was from 320 degrees at 32 knots gusting to 45 when the airplane flipped while taxiing. Could a 172 be operated in such a wind without flipping? Maybe, maybe not. It's a cinch that the technique would have to be perfect for it to work and that a wind limit should be below that value. As with weather, though, wind is free to do as it pleases regardless of the forecast, and what you see is what you get.
It is an acknowledged fact that the margins are smallest when taking off, approaching and landing, and greatest when flying along in cruise flight in good weather. Cruising puts every element comfortably within the envelope.
Even at cruise, though, an untoward event can suck the air right out of those margins. Say the engine fails on a single or one engine fails on a twin. The powerless forced landing in a single and the engine out approach to a runway in a twin are quite different maneuvers, but they have a lot in common. The margins are thin because the available energy is either none or half. Either is a bad deal. The low-speed edge of the envelope is critical in both cases and the final outcome in either relates directly to how well the pilot does at maintaining that razor-thin margin.
Margins are nice things to have. One that I always maintained is in relation to airspeed. I never ever flew the airspeed into the yellow arc, which is for smooth air only. I just never wanted to depend on air staying smooth, so when the airspeed approached the yellow I let the streak down my back of the same color prevail.