Stacked Up Over Anaconda

March 1, 2012

A battle that took place from March 2-16, 2002, in the high mountains of eastern Afghanistan, gave regular American forces their first taste of Operation Enduring Freedom’s sustained combat. This particular mission, Operation Anaconda, proved much deadlier and more difficult than its planners had expected.

Soldiers from the Army’s 10th Mountain Division dig into fighting positions after a day of responding to enemy fire during Operation Anaconda. (USA photos by Spc. David Marck Jr.)

Anaconda had been designed as an Afghan-led, US-supported operation to encircle, trap, and destroy a growing Taliban force. It would push retreating fighters into a blocking element so that they, too, could be stopped and defeated, to prevent large numbers of the enemy from melting away such as had happened the previous year at Tora Bora.

The operation was conceived weeks after the battle of Tora Bora in December 2001 saw Osama bin Laden and al Qaeda forces slip away from a mountain redoubt.

Now, the “enemy had massed for the second time,” wrote Air Force Gen. Richard B. Myers, who was Chairman of the Joint Chiefs of Staff in 2002, in his 2009 book, Eyes on the Horizon. It seemed like a golden opportunity to continue pursuit of al Qaeda leaders and fighters as the new provisional government in Afghanistan consolidated.

Special operations forces started planning. The mission was to involve the largest group of conventional ground forces of the war, so in February the SOF troops handed tactical control of Anaconda off to the regular Army, under the name Combined Joint Task Force Mountain.

Flaws in the plan perhaps began with the handoff. Up until March 2002, Operation Enduring Freedom was a model of a new kind of war. Roaming airpower supported small teams of SOF operators, Afghan allies, and CIA specialists, and there was no formal land component in place until late November 2001.

For airpower, it was easy to manage because teams were dotted all over Afghanistan. Even at Tora Bora, there had been no more than several dozen US personnel on the ground. Anaconda was designed as a different type of operation. The assault was shaped to Army doctrine as the first major ground operation of the war.

The enemy had “started to get together in a place where they could have enough mass to be effective,” said Myers in a March 4, 2002, briefing. “We’ve been following that, allowing it to develop until we thought it was the proper time to strike.”

Of course, there were multiple problems.

Intelligence slipped up. The Afghan portion of the attack collapsed. And soldiers got a shock from the level of resistance mounted by al Qaeda forces.

But the real black mark of Anaconda was the failure of Army planners to bring the joint air component into the planning process until the last minute. Operation Anaconda remains one of the central battles of the Afghanistan war because it exposed a recurrent flaw in US military operations: poor coordination between air and land forces.

The operation was “a bit of a surprise,” said retired Air Force Gen. T. Michael Moseley, who during Anaconda been the three-star air commander for US Central Command, in a 2011 interview.

Moseley did not find out a large operation was brewing until several days later. He was at a meeting elsewhere in the CENTCOM theater when Maj. Gen. John D. W. Corley called him from the combined air operations center. “I think you need to come back. There is something going on,” Corley told Moseley.

Organizing airpower had not been a major problem in Afghanistan before. SOF typically blocked out areas of operation as small restricted zones. In contrast, a regular Army force would rely on formatting most of the battle area for airspace control. The 10th Mountain Division, like other divisions, trained with air liaison officers to organize air control measures.

Yet planners breezed past these typical procedures in the haste to get on with Anaconda.

The speed with which air support could arrive was greatly affected by where the Army set the fire support coordination line or FSCL. (FSCLs are the bomb lines used to demarcate areas along the front.) Beyond the FSCL, strike sorties could seek out enemy targets.

50 Percent Intelligence

A soldier from the 101st Airborne Division scans the mountainous Afghan countryside for targets. (USA photos by Spc. David Marck Jr.)

Inside the FSCL, every close air support sortie required the highest level of control. That ensured troop safety, but also slowed down the flow of air support. Plans called for treating the entire Anaconda area of operations as “short of the FSCL, requiring positive terminal direct control and approval for strike residing with the CFLCC-Fwd Commander,” found an Air Force after-action report. Routine procedures were setting up a clog in close air support.

Air mobility planners were also in the dark about Anaconda. Brig. Gen. Winfield Scott III was the CAOC’s director of mobility forces. He could not “remember the AMD [Air Mobility Division] ever seeing the plan” for Anaconda until the system spat out an airlift requirement.

But the Army had to move between 700 and 1,000 personnel from Kandahar to Bagram for the operation. At the beginning of 2002, Bagram was a barren base with pockmarked Soviet-era concrete. Fuel was scarce, with barely enough on hand to support C-130s and Army helicopter aviation.

“We gathered up every available flying resource that we could in that part of the world,” said Corley. Aircraft dragooned into the operational support included some of the C-17s being used for a vice presidential trip to the region and Marine Corps KC-130s. Army helicopters were so short of fuel that another ground fuel bladder was delivered to Bagram, and two C-17s did nothing but service it.

“We had a tanker overhead. The C-17 would spiral up, plug in, get the gas, spiral back down, and offload gas,” Scott recalled.

Despite all this, the planners may have gotten away with the rough start were it not for a major underestimation of al Qaeda’s strength and determination.

Intelligence did not know the number of al Qaeda and insurgents in the mountains. “Before we went in there, we heard everything from 200 to several thousand,” remarked Myers at the time.

The low estimate was 168 and the high estimate, from CENTCOM, was more than 1,000 enemy personnel.

The lower estimate was used for planning and appeared in the CONOPS for the operation.

In essence, CJTF Mountain was planning for a largely unopposed operation, in which they would hold a seven-to-one manpower advantage. The situation was very different; in fact, the numbers would prove about equal.

Hundreds of enemy fighters manned dug-in positions on the high ground—above where Army helicopters would land to insert soldiers. And the fighters were prepared, with hardened defensive positions, sniper rifles, rocket-propelled grenades and mortars, and even crew-served machine guns.

“We only probably had about 50 percent of the intelligence right—locations and more importantly, the enemy’s intent, which was to stand and fight,”said the commander of the operation, Army Maj. Gen. Franklin L. “Buster” Hagenbeck, a year after the battle.

On the morning of March 2, 2002, helicopters lifted the soldiers of CJTF Mountain into the Shah-i-Kot Valley. Task force elements landed at seven blocking points to intercept enemy fighters. They immediately took fire from the slopes above.

US participants experienced scads of confusion and fought through withering al Qaeda gunfire from fighters encamped in the hillsides above, while supporting air strikes felt as if they took forever to arrive.

“What happened is … that terrain broke us up, we were fighting in squads and platoons, so you could literally be 200 yards away from a fellow platoon but you’re separated by a ridge line,” Hagenbeck said.

On the ground, the Army reacted quickly. Helicopters extracted some of the assault force from southern positions and consolidated around other soldiers. But mounting these operations under the barrage of fire led to numerous calls for close air support—far more than the two simultaneous events pledged.

Soldiers on March 14, 2002, watch a Chinook helicopter drop Canadian troops in the mountains where al Qaeda and Taliban forces were holed up in caves.(USA photo by Spc. Andres J. Rodriguez)

The Army forces had gone in very light, without their own artillery or other heavy firepower. They would be dependent on unplanned CAS when the situation deteriorated. Their main source of indirect fire support was from Apache helicopters and the air component’s fighters and bombers.

Now they would be counting on airpower. The coalition air component delivered 177 precision bombs and strafing attacks in the first 24 hours alone. There were more than 30 forward air controllers in the area, and because of the tight space, coordinators had to take extreme care to ensure bombs did not hit friendly forces—or other aircraft in the congested, multilayered airspace.

“Enemy continues to hold the high ground,” noted the Army situation report for the evening of March 2. The next day, CJTF Mountain kicked the fight into high gear and committed the theater reserve.

Moseley ordered fighters from Kuwait to beef up air support.

“We were flying missions out of Kuwait to bail these guys out,” he later said. The first two A-10s flew a five-hour commute and arrived over the battle area at sunset. Their pilots heard “two or three different ground FACs screaming for emergency CAS,” according to the Air Force report.

Soldiers on the ground “were apparently under fire with heavy machine gun and mortars. … You could see tracer fire and pockets of fire all over the place,” said one pilot. The two A-10s released Mk 82 bombs set for airburst to hit enemy troops at a mortar position.

After that attack, “the ground FAC said that all the fire they were taking ceased and that it looked like we whacked these guys out in the open. There wasn’t much movement out there anymore.”

In the air, it was a melee. Pulling up from one pass, an A-10 came within 300 yards of an orbiting gunship. Later, the A-10s were surprised when Navy F/A-18s, launching from carriers 600 miles south, dropped weapons underneath them.

The traffic convinced the A-10 pilots they were going to have to take a more proactive role in their other mission, forward air control, because they could not shuttle the hundreds of miles to Kuwait and still provide the CAS needed.

Moseley reached the Chief of Pakistan’s Air Force on his cell phone. “I want to park some A-10s at Jacobabad,” Moseley told him.

“Just tell me, are they already in the air?” asked Air Chief Marshal Mushaf Ali Mir. Moseley admitted they were.

Pakistan agreed to the emergency hosting.

Delivering more air strikes was essential, but concentrating them into an area the size of the District of Columbia was extremely challenging. The battle area was about 64 square miles and held as many as 1,400 Americans on the ground. Into that box poured an average of 253 bombs per day.

Takur Gar

A small USAF team linked with Hagenbeck’s headquarters and worked around the clock to provide a channel between the CAOC, airborne aircraft, and ground controllers.

But the worst was yet to come. Seven of Anaconda’s eight fatalities occurred on a snowy ridgeline named Takur Gar. A Special Forces team had pulled out of the ridge during hot fighting on Day 1.

Smart bombs dropped by a B-52 explode on a ridgeline during Anaconda. (Photo via

Two helicopters carrying a SEAL team attempted to reinsert back at Takur Gar early on the morning of March 4. It was important to seize the area, because the ridgeline offered a commanding view of the entire area. But above the ridge, three feet of freshly fallen snow hid hardened al Qaeda bunkers and fighting positions.

Enemy fire hit one helicopter, and as it lifted off, Petty Officer 1st Class Neil C. Roberts fell out and was killed fighting on the ground. The second SEAL helicopter returned to rescue Roberts, while an Army Ranger quick reaction force launched two helicopters from Bagram.

One of the Ranger helicopters was shot down by an RPG, instantly killing four soldiers.

Soldiers, SEALs, and airmen set up defensive positions about 150 feet from the snow-covered al Qaeda positions. The disabled Chinook was their shelter, but it made for a fat target. The troops attacked uphill, in the snow, but could not reach al Qaeda firing positions.

Close air support was the only option for their survival. Combat controller SSgt. Gabriel P. Brown was on the ground with the stranded team. “My job was to concentrate on bringing in the bombs to knock out the enemy, and I knew I needed to do it fast,” Brown later said.

SrA. Jason D. Cunningham, a pararescueman with the 38th Rescue Squadron at Moody AFB, Ga., had been on the first quick reaction helicopter. For hours he treated casualties inside the downed helicopter while taking continuous fire from al Qaeda. For their safety, Cunningham moved the casualties outside the helicopter, putting himself in the line of fire numerous times. Cunningham was fatally wounded and died on the ridge seven hours later. He was posthumously awarded the Air Force Cross for his courage under fire.

Deep into the fight Brown directed two F-15Es to his location. “We have enemy troops 75 meters away. … I need guns only,” he told them. Maj. Chris Short led the first-ever close air support combat strafing from an F-15E. On his third pass, Short’s gun delivered 100 rounds into the enemy position. “You could smell the burning pine off the trees and see snow kicking off the ground,” Brown said.

All told, Brown controlled about 30 strikes on the Takur Gar ridge that day. The survivors held off al Qaeda for 14 hours until night fell, when helicopters finally extracted the team.

The fighting continued, and deconfliction became ever more critical as the number of aircraft increased in the congested airspace above the battle.

In another twist, Afghanistan’s provisional government had reopened civil air routes mere days before the operation. Civil air traffic was flying above the battlespace, with bombers several thousand feet below and fighters streaking past at still lower altitudes.

Remotely operated MQ-1 Predators were in the mix, too. “Initially, the Predators were flying about the same altitude that we were so sometimes you’d have fairly close passes,” said Navy Lt. Cmdr. Todd Marzano, who flew several missions during Anaconda.

“That confusion and chaos could have been avoided,” said Moseley.

The good news was that bombs available soon outnumbered targets, and airpower pulled off a series of critical successes.

On a typical day, there were three or four troops-in-contact situations requiring immediate close air support. One afternoon, a single B-1 delivered 19 precision bombs on 10 different targets over two hours.

In another case, an al Qaeda mortar team fired on 10th Mountain Division soldiers for two days until it was finally destroyed by an F-16 strike and the besieged unit’s follow-on mortar attack.

On the ground the final task to complete Anaconda was to take Objective Ginger, the area surrounding Takur Gar. Objective Ginger had turned into the last holdout for al Qaeda insurgents.

The assault began with planned air strikes, which were now proceeding much more smoothly. In one strike March 8, Navy pilots tanked, checked in, and “they gave us a target immediately and then we released on that target, tanked and went home,” said Marzano.

Still, the airspace was tight. As Navy Lt. Eric Taylor set up one run to release Mk 82s on March 10, he saw “the B-52s, contrails coming in overhead the same target area getting ready to release,” he said. He knew the CAOC timed each strike, but “you knew he was coming and you knew he was going to release something on that run, so it made you a little uneasy coming through there.”

Lessons Learned and Forgotten

An F-15E takes off. A USAF pilot during Anaconda performed the first-ever Strike Eagle close air support strafing attack, in response to a combat controller request for “guns only.” (USAF photo by SrA. James Harper)

A total of 667 weapons were released on March 9 and 10. By the morning of March 10, Objective Ginger was in coalition hands. Friendly Afghan forces captured the rest of the Shah-i-Kot Valley on the morning of March 12, by which time allied aircraft had delivered more than 2,500 bombs.

Operation Anaconda ended March 16.

Tactical excellence, both on the ground and in the air, ultimately turned Anaconda into a success, but it had been a nasty shock. Those involved hustled to draw its lessons, especially as planning was under way for a new fight in Iraq.

Anaconda stood out as a reminder that regardless of technological excellence, successful combat hinges on commanders’ decisions.

“If CENTCOM had insisted that Hagenbeck build a truly joint operations plan that tightly welded the ground, air, and special operations elements, and if the command in Afghanistan had had the ability to execute a more thorough reconnaissance, our special forces and conventional troops might not have been surprised by the large, determined, and well-armed enemy,” Myers concluded.

Could it happen again? The 14-day battle has been the subject of several major reports from USAF, RAND, National Defense University, Air University, and others.

Hagenbeck publicly criticized the Air Force’s CAS speed and performance in Field Artillery, an Army publication. The Air Force initially had limited assets available for what was essentially a surprise operation, and there was a need for extreme caution as dozens of air controllers called in strikes in a small area. Nonetheless, Hagenbeck lamented the small number of aircraft that could attack at any given time and what he deemed their slow response.

Journalist Sean D. Naylor wrote a detailed book about Anaconda titled Not a Good Day To Die. War college papers and journal articles covered aspects of Anaconda. As recently as 2009, Esquire magazine published a lurid first person memoir from three soldiers who fought in Operation Anaconda.

Reminiscences of other coordination failures—such as Kasserine Pass in World War II—suggest sometimes US forces must painfully relearn the skill of air and land component cooperation.

Current joint doctrine has not permanently sewn up all the seams that plagued Anaconda.

USAF Maj. David J. Lyle studied Anaconda for a 2009 Army Command and General Staff College thesis. He found a lingering tendency to put joint task force planning first and work the components—such as air support—later.

“There is currently no doctrine available to estimate the additional bill in close air support or intelligence support if an Army unit has reduced organic capability for either self- or enemy-imposed reasons,” Lyle noted. Air and ground component commanders still need better ways of evaluating risks, especially “when logistics, geography, politics, or time limit the ground component’s ability to deploy with its full complement of organic assets,” he wrote.

Sound joint doctrine also can’t guarantee senior Army commanders will know or respect airpower.

After Anaconda, top Air Force and Army officers met to clear the air, work through the problems, and ensure air-ground coordination would be better for Operation Iraqi Freedom. The two components have integrated much better since, but Anaconda serves as a reminder that proper planning requires proactive integration.

Rebecca Grant is president of IRIS Independent Research. Her most recent article for Air Force Magazine was “Black Bomber Blues” in the January issue.