Black-Hole Approach into an Unlit Gravel Strip at Sesriem
Two pilots are ferrying a Cessna 210 from Eros, Windhoek, to a private gravel strip 18 nm south-west of Sesriem. It is 21:40 local, the moon has set, there is no cultural lighting within 40 nm and no visible horizon. The only ground references are six low-intensity battery edge lights. The 1,150 m gravel strip slopes up 1.2% toward the north threshold, with high ground 1.5 nm short of it. The pilot flying, who has 60 hours of night time in total, is on a straight-in visual approach from 8 nm at 500 ft above strip elevation and reports the lights 'sitting low in the windscreen'. The right-seat pilot reads the VSI at 500 ft/min down at 90 kt groundspeed. The pilot flying then fixates on the lights, which appear to drift left and dim. There is no approach slope indicator and no ATC.
1. Identify the visual illusions in play and explain, using the figures given, what the crew perceives compared with the aircraft's actual flight path.
Read the worked answer
Illusions. (1) Black-hole approach - with no horizon and no peripheral lighting there is no usable visual glidepath cue; the eye fixates on the only lit objects. (2) Upsloping runway - a 1.2% upslope, reinforced by high ground 1.5 nm short, produces a picture the brain reads as 'too high', biasing the pilot to descend lower than intended. (3) Autokinesis - a small cluster of point lights fixated for roughly 8–10 seconds appears to drift and dim; the reported left drift is autokinesis, not aircraft motion.
Numbers. 500 ft/min at 90 kt is a gradient of 500 ÷ (90 × 101.3) ≈ 5.5%, about 3.1°. That is a normal approach path, so the aircraft is not actually high - it only looks high. The dangerous correction is to push down to 'fix' the picture, which places the aircraft below path and toward terrain 1.5 nm short of the threshold.
Conclusion. Neither the out-of-window picture nor the moving lights can be trusted as the glidepath reference; the altimeter and VSI must become primary.
2. Set out the technique and the go-around criteria the crew should apply from this point.
Read the worked answer
Take control explicitly ('I have control'), then break the fixation by moving the scan inside: attitude, altimeter, VSI, airspeed, with only brief glances outside to keep the lights in view. Target: 3° at 90 kt is about 480 ft/min (groundspeed × 5.3). Fly that number, not the picture. Gates: a set altitude check at 3 nm and 1 nm, and a decision height 200–300 ft above strip elevation.
Go-around triggers - any one is sufficient: VSI more than 100 ft/min from target and not corrected within a few seconds; airspeed outside 80–100 kt; uncertainty about lateral position; lights not clearly identifiable by 1 nm; more than 700 ft/min below 300 ft; or the non-flying pilot calling 'not stable'.
Missed approach: full power, climb straight ahead on the runway heading, then either a stabilised overflight before any second approach or a diversion to a lit aerodrome. With unlit high ground 1.5 nm short and a pilot with 60 hours of night time, a second attempt at the same strip is not a good use of the remaining risk.
3. The right-seat pilot has said nothing so far. Analyse the CRM failure and list what should have been briefed before the approach.
Read the worked answer
A silent monitoring pilot is a classic failure mode: with a low-experience pilot flying, an unfamiliar field and heavy visual workload, the second pilot's only value is to be the instrument cross-check and the person willing to call the go-around. Silence converts two pilots into one.
Brief before descent: who flies and who monitors; the target VSI and the altitude gates; the exact go-around call ('go around, I have control'); the missed-approach heading and the divert field with its fuel implication; and a sterile-cockpit rule for the approach.
Wider judgement. A night approach into an unlit, unfamiliar gravel strip, with no slope guidance and high ground short of the threshold, sits outside a sensible risk threshold for this pilot. The real error was made before departure. The correct intervention now is not a better-flown approach but a diversion to a lit aerodrome, and the right-seat pilot should say so directly.