7. Stop 2 tells K-Car heavy fire coming from open area to his (stop 2) north. Does not seem likely area but K-Car investigates. H4 (A/S/L Hatfield) relieves E4. 2nd pair Hunters arrive. F3 positions more troops at Groot Vlei (UL 180415).
8. K-Car is talked onto area by Stop 2 (area ‘B’). Opens fire on a few (old abandoned) huts. No results. Still heavy fire at K-Car. Then K-Car sees approx. 6 ters in area ‘B’, opens fire. Ters E. 1 Frantan/.303 on target. (There were actually 11 ters in area ‘B’).
9. Hunters[1] attack target. K-Car cannon fixed. K-Car opens fire— cannon jams.
10. Pink Section troop in extra troops (now total of 16 stops). Pink 1 fired at by smallarms overshoots LZ. (Fire came from area ‘B’ prior to air attack.)
11. K-Car clears area. Eagle c/s 3 has contact with 3 ters. 3 ters killed (area ‘C’). H4 drops 1 frantan/.303 N-S.
12. 2 more ters killed in subsequent follow up. Other 120 or so ters presumably legging it to Mozambique.
This was the first use of flechettes; so OC SAS was present for yet another Air Force ‘first’.
Preparing to attack external bases
BEFORE HIS POSTING TO COMOPS, Group Captain Norman Walsh had been Director Operations in Air Staff. Never one to sit and wait for action, Norman left his office as often as possible to seek out all available intelligence. During late November 1976 through to February 77 he and Peter McLurg always seemed to be in an awful hurry as they rushed between Air HQ, JSPIS at New Sarum and SAS HQ at Kabrit. During this time, Norman came to me regularly to keep abreast of weapons developments in which he was profoundly interested.
On one of these visits in January 1977, he and Peter laid a large-scale photograph across my desk. It was of ZANLA’s main base and headquarters in Mozambique. Peter explained the layout of Chimoio Base and gave me a run-down on all available intelligence relating to it. Well over 6,000 ZANLA were known to be in residence with more trained personnel arriving daily from Tanzania and other countries. Norman warned that this was top-secret information and that my involvement was purely to assist him select the best weapons for a possible attack on the base. At that time Alpha bomb stocks were mounting but Golf bombs were not yet in production.
When he was posted to COMOPS two months later, Norman asked me to visit him regularly to keep him informed about weapons availability and run him through current project work. By that time Golf bombs were available and had already been integrated into his airstrike plans. Norman knew about the flechette dispenser system and the startling effects that could be expected from this weapon, so I suggested he consider them for use against large concentrations of exposed enemy forces, such as mass parades seen on some aerial photographs.
By June 1977, intelligence had established that there were 8,000 ZANLA in Chimoio Base and photographic evidence confirmed that the base was growing rapidly. Many miles to the north of Chimoio, inside the Tete Province, a second ZANLA base for about 4,000 CTs was also being monitored. Norman and his SAS colleagues linked this base, Tembue, with their attack plans for Chimoio. Realising that the threat from such large concentrations of trained CTs was overwhelming; they sought COMOPS authority to execute their plan.
General Walls and his COMOPS staff attended a number of presentations at SAS HQ. These were made around large-scale models of the two targets. The operational proposals frightened those who listened because they were madly daring and very dangerous. Both proposals involved relatively straightforward air attacks that were to be followed with vertical envelopment by paratroopers and heli-borne forces. This was considered absolutely essential to ensure maximum results and to seriously disrupt ZANLA.
From the outset COMOPS totally rejected any idea of attacking Tembue due to its great distance from Rhodesia; but there was a softening towards the strike plan for Chimoio. Eventually, after many persuasive presentations and much lost time, the operational plans for both Chimoio and Tembue were approved. By then it was late October 1977 when Chimoio’s numbers had risen to 11,000. Tembue was still reported to contain 4,000.
The task that Norman and his planning colleague Major Brian Robinson had on their hands was a daunting one. It was one thing to make operational proposals, but quite another to reduce them to the finest details that were so necessary to ensure effective execution.
The onset of the rainy season in November made it imperative that the attacks went in before large numbers of CTs launched into Rhodesia to take advantage of good bush cover with abundant food supplies. Had the attack been approved earlier any threat of bad weather affecting plans would not have been a factor, but now it was a critical issue and accurate weather forecasting was essential.
For some time a fundi in Salisbury had been providing the Air Force with weather information that he alone could receive indirectly from space satellites, using his own homemade equipment. Due to international sanctions Rhodesia was unable, officially that is, to receive satellite imagery of the weather patterns affecting southern Africa. How this man managed to tap into the Intelstat transmissions from Europe escapes me.
Cloud cover images were beamed down in digital form to the Intelstat receiver in Europe. Eight hours later, having been processed into usable form, the information was transmitted to a network of official receiver stations that subscribed to the service. Surreptitious interception of these signals in Salisbury obviously cost nothing but the data acquired had to be processed in a special way to get a printout resembling a photograph of cloud formations over southern Africa, as seen by the satellite. At around 10:00 every day this cloud-map arrived at Air HQ.
The cloud-maps were important to us because reports from weather stations to the north of Rhodesia were completely unreliable. Although the images we received were rather poor by modern standards, they were sufficient to warn of any major weather fronts that might affect operational planning. Nevertheless, they could not be relied upon to forecast localised orographic and thermal cloud situations that might afect long-range operations.
With the best will in the world, and using every available strike aircraft, the Air Force could not hope to produce the meaningful kill and serious wound rates we needed, because Chimoio was made up of so many camps spread over a vast area. There was no alternative but to use the very best available fighting soldiers to assault the targets immediately after a maximum-effort jet-strike.
Chimoio lay over ninety kilometres from the Rhodesian border and Tembue was almost three times that distance. However, Norman and Brian agreed that these distances favoured them in that they were both convinced neither ZANLA nor FRELIMO would seriously expect a combined air and ground attack so far from Rhodesia’s border. Until now, all combined external operations had occurred very close to the border.
The Selous Scouts’ attack on Nyadzonya Base in August 1976 had quite obviously taught ZANLA not to concentrate its forces, which is why Chimoio Base comprised so many small camps widely spread; and to a lesser extent this also applied to Tembue. Though the wide spread of targets compounded planning difficulties, it did not alter the planners’ view that ZANLA felt perfectly safe from anything but air attack, particularly with FRELIMO’s main base at Chimoio (previously Vila Pery) being so close by.
Planning the air attacks was simple enough and the selection of troops was obvious; the SAS and RLI would be used. But any idea of employing a mobile column to get a large ground force to either target was a non-starter as surprise would be impossible to achieve. Interference from FRELIMO would certainly occur early and involve heavy fighting most of the way to target, thereby giving ZANLA all the time in the world to vacate their bases. It was clear therefore that the ground force had to go in by air and be recovered the same way.
1
Hunter strike using flechette proved very successful. Total of 11 dead found in area of flechette strike. Final tally 28 CTs killed of which 5 accounted for by stops.