At that point, the MiGs ensured that FAPLA had air superiority in southern Angola. They had the advantage of holding time, or time-over-target, as they had a mere 60 to 80 km to fly to the combat zones. The South African Mirage F1s and Buccaneers, however, had to fly from Grootfontein; by the time they reached their targets, they had only ten minutes of fuel left. Hence they seldom engaged in dogfights, as their window of opportunity was too limited. The Impala jets operating from Ondangwa and Rundu had neither the range nor the armament to engage the MiGs.
With Dave Drew’s assistance, Diedies had been studying the patterns at Menongue, in Cuando-Cubango province, trying to ascertain the number of fighters, where they were parked, how they were protected and how they could be approached.
While the rest of 5 Recce was deployed on Operation Colosseum, Diedies travelled to Pretoria to make presentations to the GOC Special Forces, Major General Joep Joubert. The idea was proposed to the Chief of the Defence Force, General Jannie Geldenhuys, who agreed to the plan in principle but had to clear it with the Minister of Defence.
The operation was approved, but we still had to present the final plan and get authority for air support. This required some powers of persuasion, as it would be considered an act of sabotage – and a breach of international conventions – if a South African soldier was caught placing explosives on FAPLA aircraft. Up to that point UNITA had claimed all the successes for downing enemy aircraft.
Menongue would be the first target, mainly because fighters were operating from there against the South African forces deployed at Cuito Cuanavale, but also because it was considered an “easier” target than Lubango.
It was business as usual as Diedies, Neves Matias, José da Costa and I returned to Sawong to prepare for our next deployment. We forced ourselves to adapt to a switched routine – working at night and resting during the day. Every night we would practise navigation, patrol tactics and stalking a target.
We spent hours stooped over maps and aerial photos, working out routes, hiding places and a final approach. A photo interpreter from the Air Force, equipped with all the latest aerial photography, state-of-the-art stereoscopes and even a scale model of the target, was there to assist us.
We also used the daylight hours to study the dimensions and vulnerable points of MiG-21 and -23 jets. Using enlarged graphics of the aircraft, we determined exactly where to position the explosives for maximum effect, the idea being that the charge would cause a secondary explosion of the fuel tanks, which in turn would damage adjacent aircraft. Our reasoning was that if a charge could be placed on every second or third aircraft, we could effectively cripple the whole airfield.
We had also tested a range of blue and grey camouflage in varying light conditions. In the end we decided to use a single type of clothing for day and night. It appeared that dark blue and olive green had similar qualities in the pitch-dark conditions we were aiming for. In better-lit areas such as dam walls and power stations, a light blue or grey colour would be suitable, but, since our target area would be fairly dark, we decided on an olive drab as best suited to both day and night conditions.
For the final stalk Diedies initially wanted us to use tracksuits made of an elasticated nylon-type material, as he thought it would be better for crawling and climbing fences. However, we soon realised this was not practical, as thorns and dry grass tended to stick to the material and it got hooked on the bushes too easily.
For protection, as well as for silent movement during the stalk, we exchanged the normal anti-track covers for grey sheepskin covers over the feet, knees and elbows. At some point during the rehearsals – and actually during our first deployment – we had the woollen covers stitched to the actual clothing, covering the knees and elbows, but this proved to be a nuisance when walking. Finally Whalla-whalla van Rensburg, our unit tailor, came up with a solution: woollen covers with broad elastic bands that could be pulled over the knees and elbows, with the added benefit of pulling the sleeves tight to the body. For the approach to target we could therefore use the normal canvas anti-track covers, and don the sheepskin protection once we started the actual stalk.
For the Menongue target we decided that three of us, Diedies, Neves and I, would approach the outskirts of town, and that Neves would remain on the high ground east of the target to maintain radio comms with the Tac HQ while Diedies and I did the final penetration.
The charges were specially prepared by EMLC – moulded explosives, Torpex with an aluminium base as incendiary, and a time-delay trigger device code-named Tiller that would give us enough time to move out of the area before arming itself. Tiller also incorporated a light sensor as well as a potent anti-lifting device. To conclude the package, each device had a tube with two chemicals. When squeezed, the chemicals mixed and formed a quick-drying, potent epoxy glue – the so-called Vernon Joynt glue, after its inventor, the ingenious Dr Vernon Joynt.
To carry the explosives to the target, we designed a pack containing compartments for nine charges, each of which could be reached without having to take the pack off – except the one in the centre of the pack, which could not easily be reached by hand. This pouch we stacked with one day’s food and our E&E kit.
Water was another challenge. In order to be agile and flexible for the stalk, we preferred not to carry water bottles in kidney pouches on the sides of the body. Thus a full water bag was attached to the centre of the rucksack. The bag had the added advantage that the water would not make a noise, as it would in a half-filled bottle, while we were on the move.
Our preparations were put to the test when, during the last two weekends before our deployment to Rundu, we conducted full dress rehearsals at Hoedspruit. There we recreated the conditions we expected at the target – the distance to approach to the perimeter fence, a dark moon phase and the distance to cover on the hardstand to reach the aircraft.
The SAAF had been requested to provide us with all necessary support, and so one of the fortified hangars was left open, while four Mirages were parked outside. The base security squadron had been put on alert for the weekend. As far as they were concerned, the Recces would be doing an exercise against them and they would go all out to detect us.
The three-man team crossed the perimeter fence undetected. We lay up inside the base for a day, and then donned the sheepskin covers and infiltrated to the hardstand where the Mirages were parked. It was a pitch-dark night, and we found that we could stalk the “enemy” – our own Small Team colleagues and some counterintelligence personnel from HQ in Pretoria guarding the planes – to within a few metres before we were detected. We practised the stalk, fine-tuning the posture and adjusting the equipment until finally, in the early hours of the morning, both Diedies and I could stalk to within touching distance of the sentry.
The dress rehearsal at Hoedspruit gave rise to another ingenious invention: a harness we designed for the night-vision goggles. The normal neck strap was no good when you had to crawl or climb. The new harness consisted of a neck strap with the addition of an elastic band around the body, which would hold the goggles tight to the chest. In this way the operator could use both hands without having to worry about the night-vision goggles swinging loose. When he wanted to use the night vision he simply had to pull it away from the chest, while the elastic would tug it back into position once he let go. Also, to prevent the greenish glow from the night vision reflecting on the face, we had a piece of sheepskin moulded around each eyepiece.