Controlled Flight Into Terrain

‘Controlled Flight Into Terrain’ (or CFIT) is the number-one cause of fatal aircraft accidents in the UK. That is to say, the pilot has not lost control of his aircraft – he is still completely in control – but the aeroplane is destroyed by a collision with the ground. How can this happen? Far and away the most common cause of this sort of accident is where the Pilot flies into a cloud* at low level, and that cloud just happens to have a solid centre. So, something like a mountain, a hill, a radio mast, that sort of thing. Pilots have a name for that sort of cloud; we call it ‘cumulogranite’. Invariably, the aeroplane loses such an argument, but, of course, the simple answer is to avoid flying into cloud at all, never mind at low level. My Pilot’s licence does not allow me to fly in cloud anyway, so it’s a no-brainer.

So, for me, a CFIT accident should never happen.

Or so you would have thought.

Let me tell you the brief story of a rare ‘hairy moment’ I managed to contrive a bit back, that could have become a CFIT but for my decisive action.

While flying one of the Piper Tomahawks from my local Flying Club in Devon, I decided to perform a ‘Practice Forced Landing’, or ‘PFL’ for short. As described in this article, it’s where you pretend that the hamster has died** and the engine has therefore failed, so you need to practice the unpowered glide down towards a suitable field. But you don’t actually land there, as you would do in a real emergency; no, once you think you’d have made it, you open the throttle again and climb away.

Now, I have only been flying from this airfield in Devon for a year or so. For the previous few years, I have been flying over the gently rolling, undulating Cornish countryside. The Devon countryside is similar, but we do have at least one giant ridge, called Haldon Hill, which rises several hundred feet out of the western side of the Exe Valley. Oh and there’s the huge, high plateau of Dartmoor as well, but that doesn’t figure in this story. Most of the time, though, Haldon Hill is not a problem because we fly at 2,000 feet or higher, so the Hill is well below us.

But for a PFL, by definition, you do go down low and fly pretty close to the ground. You are pretending you’re going to land, after all.***

So, on this particular occasion, I ‘failed’ the engine at 2,000ft at a randomly-chosen time and chose a field a few miles west of Starcross on the Exe estuary (‘randomly-chosen time’ because you don’t look for a nice big field and then fail the engine once you’ve found one; in real life, the engine would likely fail with little or no warning and you wouldn’t have that luxury). A nicely-sized triangular field on a slight uphill slope (this helps slow the plane down once you’ve landed) and directly into wind, and with a farm next to it. In real life that would have been the perfect place to land. And, also noting that Haldon Hill was well clear a good few miles away to the west so that should be fine; there was plenty of space for me to climb away so I don’t hit it.

So, I conducted the PFL drill and got down to about 300ft. I wanted to go a little lower than normal – usually no lower than 500ft above ground level – to see what an open field looks like from a lower level. Having decided that, had that been a real forced landing, I would have survived, I selected full power and began the climb away. Looking up from the field, well I can’t say I went into a panic because I don’t do that, but right there in front of me, less than a mile away, was a bloody great big treeline – a ridge with loads of trees on it – and I was looking up at it! The top of the treeline was a good 250ft above me…

You see, because I had descended for the PFL from above the height of Haldon Hill, I had failed to see the treeline in the foreground and a mile beyond ‘my’ field, because it was well below Haldon Hill. It was down in the ‘ground clutter’, as it were, and as I was concentrating on the field and the landing procedure, I hadn’t seen the ridge under my ‘exit route’. I found out a few days later – by finding ‘my’ field on a map – that the treeline is a medium-sized copse near a wood called ‘Mamhead Big Wood’. Maybe it was called ‘Mamhead Little Copse’ or something. Well, whatever it was called, it certainly looked like it was living up to the larger name right now, Mamhead Big Wood or not, and not only that but it was above me! and it was also getting closer. With less than a mile to run, low on airspeed because I was just recovering from the descent and getting rid of the drag flaps and also nose-up as well (which reduces the acceleration), and not yet at best climb speed, I knew I was going to have to do something pretty decisive. Low on airspeed, altitude and ideas, this is a situation we Pilots call ‘tumbleweed’ for obvious reasons. No way was I going to gain sufficient height to clear ‘Mamhead Little (but getting bigger) Copse’ in the time available.

Adopting the only course left open to me, I decided to turn gently right – again, in a climb, your turning is restricted too because turning uses ‘lift’ from the wings, and when you are climbing you are using most of that already – so the gentle right turn brought me northwards and parallel to the ridgeline. With that terrain feature now to my left, I now had plenty more space to complete the recovery manoeuvre. CFIT avoided.

Map showing my PFL flightpath. Dotted line shows the gliding descent, ‘my’ target field is the triangular one marked with the blue ‘X’, and the climbout escape path is shown as a curving solid line. Arrowheads give direction of travel. ‘Mamhead Big Wood’ visible to the left of the map, but actually it is on the reverse slope into the next valley; the problem wood was that smaller copse to the east. Look at those contour lines; each one of those is 5 metres more altitude…but also see how the land drops away again west of the ridge through Mamhead Big Wood. Scale: Large blue grid square – 1km

So, I learned from this that I should only really do a PFL in properly rolling countryside; I know now that the area between the Exe estuary and Haldon Hill is a tricky complex of ridgelines, slopes and forests which is not good terrain in which to practice forced landings. I have no doubt I would get a plane down in that area safely should the real thing happen, but for practice? Not good, because the exit routes are potentially not good. When you’re trained to do PFLs, you’re trained to look out for certain types of field: fairly level or with a slight upslope; big enough to land in with enough space to stop before you hit the far hedge; into wind and hopefully not into the sun (but the wind direction is more important); hopefully a reasonably smooth and firm ground surface; preferably near to civilization so you can get a cup of tea while you wait for the AA to come out; and no obstructions like trees or power lines on the approach path. But nobody ever mentioned the escape routes. The vast majority of the time, you are going to be doing these sorts of things for practice only, so it’s not just a matter of ‘would I have got into the selected field safely’, but also ‘can I get out again once I have completed the PFL exercise?

The day a Pilot stops learning is the day he should stop flying. I’m glad to say that I learned from this event; I got in some good practical and real-time decision-making, and I lived to tell the tale. I wouldn’t say that the event was necessarily dangerous, but it was what we might call ‘marginal’; not much room for error and very much dependent on my skill, experience and knowledge of my aircraft in order to resolve it satisfactorily.

I’m probably going to drive over and have a closer look at ‘Mamhead Big Wood’ and ‘Mamhead Little Copse’ sometime. I would say that it’s in order to see what it looks like from the ground, but I would think that I can make a pretty-well informed guess about that already, don’t you? 😉


*The number two cause of fatal aircraft accidents is ‘loss of control in IMC’ (‘IMC’ meaning ‘Instrument Meteorological Conditions’ or, again, cloud), that is, the Pilot gets into cloud, gets disorientated, and loses control of the aircraft. Again, very rare and easily avoided (because we’re not supposed to fly into cloud for that very reason), but still worth bearing in mind.

**Everyone knows that light aeroplane engines are powered by a hamster running on a little wheel, under the engine cowling; this makes the propeller go round very fast.

***People sometimes ask me “How low can you fly in those aeroplanes of yours? How close to the ground can you go?” To which, of course, the only correct answer is, “All the way down. It’s called “landing” ‘ 😉


Header image shows Piper PA-38 Tomahawk G-RVRR, callsign ‘Romeo-Romeo’, one of the two Tomahawks I fly.

3 thoughts on “Controlled Flight Into Terrain

  1. One of these days, you may want to get an IFR rating – it gives you more options in dealing with those clouds. 🙂 Back in the day, when my dad still flew, you were allowed (at least here in the US) to do a controlled spin and recovery in a Tomahawk. Crazy.

    1. Yeah, I might do that one day, but it’s the Night Rating for me first. I’m going to do that this next winter. Here in the UK, unlike in the States, the Night Rating is not a part of the standard PPL training syllabus.

      Spinning – yeah, I have spun gliders, and a Cessna 152 as a requested part of my training (I asked to be shown spins and be allowed to practice recovery). But the Tomahawk is a tricky aeroplane in the stall/spin regime; it is actually designed to have more ‘difficult’ (read: vicious!) stall/spin characteristics than other training aeroplanes. We are not allowed to spin them solo and I personally think that a lot of instructors are afraid of spinning them, so it looks as if I won’t get chance to spin one. I have watched YouTube videos on the spin characteristics of the Tomahawk and it looks pretty straightforward if you know what you’re doing; the main thing is that as the aircraft is recovering the rate of spin actually increases temporarily and *then* it recovers, and a lot of pilots don’t know about that and so abandon the recovery attempt. Quite how they manage to recover eventually I don’t know. The Tomahawk was designed that way so that student pilots actually had an aeroplane that wasn’t easy in the stall, unlike the Cessna 152 or the Piper PA-28 Warrior. Personally I think the Tomahawk isn’t a good training aeroplane precisely because of this. You need a trainer to have docile handling characteristics; yes, spinnable with the correct control inputs but the Tomahawk…almost invariably she will drop a wing quite harshly at the stall, with an incipient spin not far behind. I am very, very wary during low-speed handling in the Tomahawks because I know what they can do. I give the aeroplane a respectable margin over the stall speed at all times! Conversely, you’ve really got to try hard to spin a Cessna 152; they don’t like to spin at all and the stall regime is really gentle and easy to deal with.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.