Two U.S. Air National Guard F-16 pilots face disciplinary action after mistakenly bombing Canadian troops in Afghanistan in April 2002, killing 4 and wounding 12. The incident may carry important implications for U.S. choices of tactical air support systems. The Canadians were in a no-fire zone, but the pilots' rules of engagement freed them to attack in the event they were fired on. That caveat is virtually standard. Moreover, pockets of Taliban loyalists continue to resist the new Karzai regime and its coalition supporters. No part of the country can be considered entirely free of such loyalists. It would seem to follow that not only is self-defense permitted, but attacks on those firing on U.S. aircraft are preferred to the alternative, which would be to climb out of light antiaircraft range. Any U.S. tactical airplane flying over Afghanistan is, in effect, on an armed reconnaissance mission.
The Canadians were conducting live-fire training in the form of an antitank exercise; they were not firing into the air. Moreover, the U.S. command had been notified that these exercises, which might seem from the air to be hostile action, would be taking place. If the pilots are not considered culpable, then one must wonder whether the chain of command was capable of taking account of the Canadians' report of their exercises, and of disseminating that claim effectively. How well can a pilot be made aware that a small patch of territory over which he passes, and with which he may not be particularly familiar, is a special no-fire zone?
Supposed evidence of pilot culpability is that the air controller of the F-16 that actually dropped the bomb called out "hold fire" after the pilot asked to use his cannon. The pilot then rolled in and dropped a laser-guided bomb. It is not clear whether he understood the "hold fire" command as merely to hold cannon fire for tactical reasons. All of this suggests that the problem was the pilot's understanding of the situation in a context possibly far removed from the formal rules of engagement. It was relevant that the Taliban had shown a disconcerting ability to regroup during Operation Anaconda; they were hardly altogether finished as a fighting force. Had the victims on the ground not been allied troops, the pilots likely would have been praised for their initiative in interpreting ground fire as the presence of the enemy.
The Canadian official report observed that if, as the pilots claimed, they were responding to what they thought was antiaircraft fire, they should have climbed away as soon as the bomb was released, to avoid any additional fire. Instead, they remained in the area. Such behavior would be consistent with the view that enemy antiaircraft fire was an indication of the presence of enemy forces, hence an appropriate target. The pilots' aim would have been to wipe out the enemy force that had revealed itself. Incidentally, this view of the incident would explain why a pilot dropped a bomb even though he saw weapons being fired at ground targets; he apparently mistakenly imagined he was watching a Taliban ambush.
One question is whether the mission itself was so long that pilot fatigue could not have been avoided, whatever stimulants the two F-16 pilots took. The F-16s were flying from a base in an unnamed Gulf country, and the attack came far into what was described as a long mission. For decades, the Air Force has maintained that modern tanking techniques give its tactical aircraft unlimited range. Some of its supporters have gone so far as to claim that such techniques make aircraft carriers, and even foreign bases, obsolete. The issue of pilot fatigue has not been raised. Perhaps in past wars pilot judgment was not as critical. In Afghanistan, however, almost no coalition troops have died at the hands of the enemy. Friendly fire has become a major cause of casualties.
At the same time, because the enemy wears no uniform, pilots and ground troops constantly are on the lookout for enemy fire as the main means of distinguishing enemy forces. Moreover, because the enemy is interspersed with the friendly population, it is difficult to designate safe zones or absolute no-fire zones.
Other things could have been done. The vital role of ground control suggests that pilots generally cannot associate what they see on the ground with positions mentioned in their briefings. The F-16s were, at least in theory, under direct control of an airborne warning and control system (AWACS), but it seems unlikely that the AWACS crew was monitoring their position to see where their weapons might be going. Instead, the role of the AWACS was probably to direct the F-16s as required to attack targets identified by ground controllers or by other sources.
It would seem that the only solution to this kind of problem, aside from reducing pilot fatigue, would be to provide the pilot with a map of friendly and presumed enemy force locations, presented in such a way that it automatically is associated with what the pilot actually sees through his canopy. Many Air National Guard F-16s, for example, are fitted to receive and display the Situational Awareness Data Link (SADL), which connects with the Army's Enhanced Position-Location Reporting System (EPLRS) data link. Another possibility would be to use Link 16, which was conceived as a close-air support system. Either solution would have provided a pilot with enough situational awareness to reduce the chance of an incident like the one that killed the Canadians. Even that might not have been enough. For example, a pilot aware of the presence of the Canadian force might still imagine that he was watching a Taliban attack on them. The degree of misinterpretation probably would be related to how fatigued the pilots were.
Mine Clearance Progresses
The Royal Norwegian Navy is demonstrating an interesting example of network-centric warfare, this time for mine clearance. The great problem in mine clearance is that a minehunter must examine minelike objects one by one, deciding which is a mine (and dealing with it) before proceeding to the next. Minehunters are expensive, not least because they have to be silenced acoustically and magnetically so they can approach mines without setting them off. Navigation must be precise, so that each minehunter can tell others which mines she has or has not neutralized. The antimine sensors also are quite expensive, because small mines must be distinguished at considerable range.
The new Norwegian approach is to place the vital sensors in a small underwater unmanned vehicle (UUV), Hugin, which propels itself through the supposed minefield. Hugin detects mines and brings its detection information back to a mother ship. It surfaces periodically to use its radio, both to transmit data and to locate itself relative to the mother ship.
The mother ship controls an antimine torpedo, Minesniper. She has no need to localize mines, because Hugin already has done that. Moreover, the mother ship has a plot showing where those mines are. Because Minesniper is command-guided, it can be steered to the mine. If localization is not perfect, Minesniper has its own onboard sensors. Because they need not detect a mine at any great distance, they can be inexpensive. The sonar on the mother ship need only be able to track Minesniper so that the weapon can be guided properly; it need not be anything as elaborate as the usual minehunting sonars. The Hugin/Minesniper system is being tested on board the minehunter.
The system is network-centric. Network-centric means that weapons are fired with reference to a large-scale tactical picture produced by multiple sensors, some of them offboard. Current minehunting is platform-centric, in that action proceeds from detection by a ship's own sonar. Even stand-off systems are platform-centric; they merely move the sensor offboard. A key advantage of the net-centric system is that the sonar on board the mine countermeasures ship need only track a transponder on the kill vehicle, commanding it to coordinates already identified by the remote sensor.
The U.S. Navy has gone partway down this path. It has developed a surface ship-launched minehunting UUV, the Remote Minehunting System (WLD-1), and also a submarine-launched minehunting UUV (LMRS, or BLQ-11). Each system can, in theory, map a minefield. If the threshold for mine recognition is not set very high, there will be many false positives. That will be acceptable, however, if the objective is to identify likely fields simply for avoidance. That is the current position. The next step would be to provide a map good enough for the field to be attacked. If the map were good enough, it might be possible to deliver mine-killing munitions by air. The munitions would, of course, have to be able to get to designated underwater positions, which probably would entail some sort of inertial navigator.
The network-centric approach seems to be the first really new idea in mine countermeasures to have emerged in many years. It seems to be well worth pursuing, as we badly need some way of rapidly clearing fields. Otherwise, mine clearance will be a dead giveaway of where we plan to land Marines. The sheer length of time needed to clear fields offshore will give an enemy time to set up beach defenses we may find difficult to counter.