April 13, 2007

Statistics: close-air support in Iraq and Afghanistan, 2004-06

A little known fact about the use of airpower in both Iraq and Afghanistan is how few munitions are actually being dropped these days.

CENTAF (Central Command Air Forces) have regularly released daily and weekly figures on Close Air Support (CAS) sorties, and munitions dropped, for several years. The most recent is here*. These are missions carried out by American and allied air force jets, over both Iraq and Afghanistan. They do not include helicopter missions, including attack helicopters. The munitions counts also do not include engagements with cannon, or unguided air-to-surface rockets. And obviously artillery engagements are not tallied here either. This is just the bombs and guided missiles dropped or launched from military aircraft.

Still, that's a pretty big number, right? How many bombs do you think have been dropped on Iraq and Afghanistan in the last three years?

I'll save you the adding of over 1,000 daily summaries: would the answer 2,950 surprise you at all? Because that's what it is.

Here's the chart:

CAS munitions

As you can see, over half of the total munitions that were expended by Coalition air forces in the last three years were over Afghanistan in 2006. The other weapons drops have been remarkably limited. Almost all of these munitions would have been precision weapons, either GPS- or laser-guided, dropped from a combination of B-1B Lancer bombers and various air force and naval-based fast jets.

What's perhaps most remarkable is the number of CAS sorties that are being launched for every drop of one munition. To clarify, a CAS sortie is counted every time a plane takes off with ground attack weapons loaded, either to loiter to wait for a target, or to deliver against a specific objective. A total of 71,237 CAS sorties have been launched by Coalition air forces in the 2004-06 period. About a third of these were over Afghanistan, and the remainder over Iraq. Many of these would have engaged with other weapons like cannon, as mentioned above. But the number of missions, relative to the number of actual large munitions dropped, is revealing:

CAS munitions 2

Now assume for argument's sake that each sortie that dropped a munition dropped only one. (In actual fact, of course, each B-1B sortie can potentially launch many more than that**.) That would mean that, at a minimum, in both theatres of war over the last three years, over 40 attack planes have gone out fully loaded and returned fully loaded, for every bomb dropped. The exception was Afghanistan in 2006 where, again assuming one bomb per plane, about one in six sorties found a target. Once one factors in the number of bombs that could be carried and dropped by different planes, those numbers would actually be much higher.

The military doesn't seem overly concerned about this ratio being a failing of any kind. Indeed, the troops on the ground probably appreciate having the bombers on call regardless of whether they are needed. However, it certainly does indicate that neither Afghanistan nor Iraq have been particularly "target-rich" for the air forces involved.

There's lots of reasons a plane would not drop a bomb, even if it had a potential target in mind, as well, ranging from mechanical errors, to avoiding friendly fire possibilities, to the target being taken out by other means, such as the local artillery unit. Legal and other targetting issues are also no doubt playing a factor.

How do these numbers compare to other bomb-dropping rates, historically? Well, in both Vietnam and World War Two, U.S. air forces combined dropped an average of over 4,000 munitions per day, outstripping the combined Iraq/Afghanistan total for the past three years by a significant total, but that's probably not a fair comparison. Sticking to the precision era, one comes up with figures like these:

CAS munitions 3

Recall that both these air wars were criticized at the time for being rather desultory, compared to, say, the 1991 Iraq war. Learned military commentators said in both cases, up to the point both wars ended favourably, that air power was not being used in sufficient quantity, since only 300-odd munitions were being dropped per day on a country the size of Serbia/Afghanistan. Well, right now Coalition forces are dropping roughly 1 per cent of that on Afghanistan and Iraq combined each day.

Which is not to argue for more bombing. Especially in Iraq, this does not appear to be a war which more bombs would help. However, it's probably worth keeping in perspective.

As mentioned here previously, aerial weapons in the precision era seem to kill civilians, as opposed to their hostile presumed targets, at a rate of about 1 civilian for every 10 munitions dropped, no matter how careful or careless the launchers are. The exact ratios, as near as historians can determine, were about 0.02 unintended fatalities per weapon in Kosovo, and about 0.07 in Afghanistan.

If one were to cross-reference a similar ratio with the numbers of bombs dropped in Iraq and Afghanistan over the last three years, even with a liberal error bar one would come up in each case with a total number of civilian fatalities due to bombing in each country in at most the low hundreds: far fewer than have been killed, for instance, by insurgent suicide bombers in both countries. This does not mean that Allied forces are not accidentally killing larger numbers of civilians in other ways: attack helicopters, artillery, small arms fire at checkpoints, all must be taking a regrettable if unintended toll, as well. But it does suggest that the aerial contribution to the civilian death tolls in Iraq and Afghanistan in the last three years cannot be a major component of, say, the 4,000 total Afghan deaths reported by the UN and Associated Press last year, or the 600,000 excess deaths in Iraq since 2003 recently estimated by researchers in the Lancet.

*So, for instance, on April 4 of this year, there were 31 close-air support missions over Afghanistan, and 55 over Iraq, in which a total of one plane (an RAF Harrier dropping a laser-guided bomb in Afghanistan) was the only engagement that used a munition.

**In Afghanistan in 2001, the strategic bombers of the 28th Air Expeditionary Wing, predominantly B-1Bs, dropped an average of 17.7 munitions per sortie.

Posted by BruceR at 01:31 PM

Haroon Siddiqui: dishonest on casualties

Haroon Siddiqui in the Star today:

"The vulnerability of the Afghans is reinforced by NATO air bombings, 2,000 last year, which killed about 4,000 people."

This sentance is fundamentally dishonest. The 4,000 figure comes from the UN and Associated Press aggregate fatality counts for 2006, which nearly agree with each other. As discussed here, that estimate was of all deaths in Afghanistan from the war, not just from bombings, including insurgent attacks. A reasonable estimate is that over 2,000 of those were deaths of either Afghan police or military personnel or insurgents, and the majority of the remainder against civilians were from insurgent attacks, as Human Rights Watch had concluded previously. (Suicide bombings alone killed over 200 civilians.)

The bombings number he uses, meanwhile, relates to munitions dropped: 1,770 by Central Command figures in 2006. Now, historically, air forces have never seen the average number of civilian deaths per munition go much greater than or much lower than about 1 civilian death per 10 munitions:

*Afghanistan in 2002 saw 17,500 weapons dropped, about 10 times this number, and is believed to have killed about 1,300 civilians.

*Kosovo saw 23,000 munitions dropped, and about 500 civilians killed.

*The 1991 Gulf War saw 250,000 munitions delivered, and about 3,200 civilians killed.

One might argue, of course, that an insurgency increases the dispersal of targets among the civilian population, which might change this ratio upward, but most of the NATO air attacks in 2006 were in remote rather than urban areas, which would tend to act on the statistics in the other way.

It would therefore be at least a historically grounded hypothesis, without reference to the media coverage, to argue that NATO air attacks must have inevitably killed approximately 200 Afghan civilians in 2006 (+/-100). Two hundred too many, certainly. But even if you double it to be safe, Siddiqui is still almost certainly lying by an order of magnitude.

PS: Printing the allegation that Canadian troops are acting "with racist disregard" in Afghanistan, without substantiation of any kind (other than the non-statistics above) is a calumny that one would once have hoped would be beneath Canada's largest-circulation newspaper, as well.

It would be much easier to handle the principled opposition to the Canadian military involvement in Afghanistan if it could once in a while act as it it were principled.

UPDATE: You know, Opposition defence critics like the NDP's Dawn Black or the Liberals' Denis Coderre would not be so derided by the pro-military crowd if they, you know, once in a while, took on the excesses of anti-military press hyenas like Siddiqui and Robert Fife with a principled letter to the editor, or something. The fact they've never, ever done that, no matter what the antiwar talking heads say, speaks volumes about their own integrity.

Posted by BruceR at 12:06 AM