For this new role, there were three essential changes to be made. Firstly, an SAS Tactical HQ would have to base permanently inside Mozambique next to a runway that could support Provosts and laden Dakotas. Secondly, patrols would have to operate in groups of only four men (known as a stick); a limitation imposed by the lift capacity of an Alouette III carrying a machine-gun and gunner. Finally, SAS sticks would have to operate six weeks in the field with only ten days’ leave between deployments.
In September 1973 Brian got his way and an SAS Tac HQ was established next to a scruffy little airstrip at Macombe. Macombe was a Portuguese aldeamento on the south bank of the Zambezi River lying midway between the north-flowing rivers Angwa and Musengezi. Two Provosts and two Alouettes positioned forward from Musengezi Mission to Macombe on a daytime-only basis, as SAS could not guarantee their safety at night.
From the outset the SAS was entirely successful in this new style of operation but the RAR deployment to Macombe was withheld for months. So any advantages given by the natural Zambezi River obstacle and the disruptive effects of the SAS operations to its north were largely lost. Nevertheless SAS disruption of ZANLA’s supply lines greatly assisted in relieving internal stresses.
I visited Brian at Macombe when he first moved there and passed him all the information I had gained from my recce work north of the Zambezi. Again I suggested to him that I should continue visual recce to continuously update the information and provide him with a daytime radio link to his troops. Brian would not entertain the idea. He continued to insist that, except in emergency, the SAS areas of operation should be free of any air activity. Contrary to Brian Robinson’s thinking, mine was that aircraft on visual reconnaissance should be a permanent feature over the entire Tete region, both within and beyond the SAS areas of operations. I contended that it was essential to provide the SAS with fresh intelligence upon which to plan, thereby saving lengthy ground recces to find the best sites to monitor, ambush or attack.
Over months to come I would build up information on civilian locations, routes and terrorist camps covering large areas beyond those in which the SAS were active. When SAS operations switched to areas I had covered, as much as six weeks previously, Brian reported that my information, though correct, was out of date. Not surprisingly the men on the ground ignored much of the air recce intelligence that had been so painstakingly put together. But one thing they helped clarify was that I had been accurate in my assessment of terrorist bases, as opposed to civilian camps. What had helped me distinguish between these sites, which looked much the same from height, were female toilet paths leading out of civilian locations.
I was so frustrated by Brian Robinson’s attitude that I opted for air-controlled strikes on fresh terrorist bases. The Air Force never expected the successes that could be achieved by a pure SAS ground attack or those involving air strikes in conjunction with the SAS. Nevertheless, we knew that FAC strikes would demoralise and disrupt FRELIMO and ZANLA in areas unattended by SAS. In too many respects our Air Force had been forced to act independently in like manner to the Portuguese, which we had criticised so strongly—but there were advantages to be gained. I continued recces north of the Zambezi, always keeping well clear of the SAS and taking time before calling for jet-strikes. Once these got under way we developed procedures that required close understanding and co-operation. It took a few calls before FAC-controlled airstrikes became a well-honed routine.
As mentioned earlier, jet-strikes in direct support of earlier SAS operations had been very hit-and-miss affairs that improved little over time. For the men on the ground, particularly in flat terrain, there were considerable difficulties in judging the exact point of a jet-strike because bush and rough ground usually intervened between the GAC location and target. This meant that corrections had to be assessed from the sound of exploding weapons or from dust rising above the bush some seconds after each strike. FAC did not suffer these difficulties.
The cost of live strikes on terrorist bases was naturally much greater than on routine weapons training, but routine training was conducted on a range that every pilot and navigator knew intimately. Unlike calls to live targets, routine training required no hurried navigation planning or the split-second timing involved in co-ordinating fast jets with a slow-flying FAC aircraft.
Perfect timing for maximum surprise followed by slick and accurate placement of weapons were all essential ingredients considering the speed at which terrorists could run clear of their bases. The value gained by jet crews scrambling to meet FAC in remote areas for strikes on never-before-seen targets was immense, and it prepared us well for the future.
With every target’s grid reference I also gave an IP (Initial Point) and the exact time of my marker strike. The jets planned to fly directly to the IP—a clearly identifiable feature out of sight and hearing range of the target itself. Formation leaders worked out the heading and precise flight time from the IP to target so as to strike no more than three seconds behind my target markers. I too would have an IP point from which to move in orbits towards the target.
Canberras could not strike anything like as quickly as the fighter-bombers because the bomb-aimer required time to pass heading corrections to his pilot which necessitated quite a large stand-off distance at the moment of mark, so they were never used alone for FAC work against fleeting targets. In addition to the long run in, the flight time of bombs exceeded 10 seconds and shrapnel settling time limited following Canberras to long intervals between individual strikes. Consequently Hunters or Vampires usually preceded Canberras. Strikes initiated by Canberras with fighter-bombers following became commonplace in later years when the bombers made their attacks in formation. However, most early FAC work was with Hunters only—usually four of them.
When the lead pilot checked in about five minutes before reaching the IP, we would synchronise watches and confirm individual running times from IPs to target. I would be told what weapons were being carried and the sequence in which they were to be delivered. This gave opportunity to suggest how the weapons should be distributed on the line of attack, relative to my target markers.
Formation leaders normally planned to arrive at IP with as much as three minutes to spare. This was useful as it allowed me to make a slow approach onto my own IP in a series of orbits that the enemy on the ground recognised as my normal flight pattern. Even when out of sight, the sound of a piston aircraft indicates whether it is flying on a straight line or orbiting. In orbit the sound varies from highest frequency when approaching to lowest frequency when departing. For reasons of surprise, I needed to ensure that this continued right up to the moment that I turned in to fire my markers.
For Hunters covering ground at the rate of seven nautical miles per minute, or Vampires and Canberras at five nm per minute, changes in wind velocity seldom affected their timing by more than parts of a second. However, the same conditions could make an enormous difference to the slow Trojan, which is why my own IP was always relatively close to target.