What AQAP’s Operations Reveal about Its Strategy in Yemen

April 23, 2015

For special access to experts and other members of the national security community, check out the new War on the Rocks membership.

The recent takeover of Yemen’s fifth largest city of al-Mukalla by al-Qaeda in the Arabian Peninsula (AQAP) highlights the growing strength of the organization. While AQAP has certainly taken advantage of the more chaotic environment as a consequence of the Houthi’s war in the south and the Saudi air campaign, the group has in fact been gearing up its own overt military campaign since last summer. Therefore, even if there is an eventual ceasefire between the Houthis and the Saudis, AQAP will continue fighting and operating on its own terms.


Starting in late July 2014, AQAP made a concerted media effort for the first time to actively report and take credit for its military operations on an almost daily basis. This differed from its past pattern of only commenting on large-scale operations. In part, AQAP did this to bring attention to its new military campaign, two years after it had been kicked out of southern cities by the Yemeni military and local popular committees after governing from the spring of 2011 to the summer of 2012.

As part of this new media effort, AQAP created different Twitter accounts online to push its content, one of them being a news feed called Akhbar Ansar al-Sharia fi Jazira al-‘Arab (Ansar al-Sharia in the Arabian Peninsula News; the name of the feed is derived from a period in 2011 and 2012, when AQAP controlled tracts of territory and adopted the name Ansar al-Sharia in Yemen to circumvent perceptions of toxicity with the AQ brand). This feed has been AQAP’s key mouthpiece for releasing information on its military activities since early August 2014. Through April 21, AQAP has claimed responsibility for 374 attacks, with the vast majority against Houthi (224) and government forces (147).* Therefore, while AQAP has certainly taken advantage of the recent chaos and vacuum created by the Houthi attacks in the south and the Saudi air campaign, the organization had already been involved in a sophisticated military campaign. In many ways, the group is now just exploiting a change in conditions, which will allow them to thrive even more in the same way The Islamic State was able to in Iraq in the lead up to its takeover of Mosul almost a year ago.

AQAP’s Operations

AQAP’s modus operandi is remarkably dynamic. But while the group’s target selection, tactics, and geographic concentration appear fluid, by analyzing its attacks since August particular patterns can be discerned, which themselves offer an opportunity to assess not only the magnitude but also the nature of the threat AQAP poses to security and stability in Yemen.

AQAP operations in Yemen by governate (click to enlarge)

In August, AQAP undertook 32 attacks, 27 of them against government positions, personnel carriers, and specific individuals. Despite this homogeneity of targets, AQAP displayed a modest ability to hit such targets in different parts of the country, illustrating that prior to the beginning of this campaign it had infrastructure in place to carry out such attacks. On top of conducting 15 operations in the central districts of Hadramout governorate, a vast area encompassing the eastern half of Yemen, AQAP struck six targets in Lahj, a southern governorate bordering Aden, five in the central governorate of Shabwa, and three in the northern governorate of al-Jawf. AQAP operatives also utilized a diverse set of guerilla tactics, including 11 IEDs, eight hit-and-run ambushes, five assassinations, and five offensive assaults, to hit 15 government positions, six government vehicles, and five state officials.

September saw the number of AQAP attacks drop from 32 to 12 — the organization’s quietest month between August 2014 and April 2015. However, September is a watershed month in that AQAP targeted Houthis six times, the same number of times the group attacked government forces. This 50-50 split marks a significant shift from the previous month, when Houthis were targeted only four times compared to the 27 operations against government forces. Additionally, AQAP’s geographic reach continued to expand, including an operation in the Houthi-dominated Sa’da governorate that borders Saudi Arabia and four in Sana’a, one of which targeted the American Embassy. The type of weaponry and targets remained constant in that a variety of ambushes, IEDs, assassinations, and offensive assaults were deployed to hit both government and Houthi forces.

AQAP conducted 43 attacks in October. Approximately 21 operations were undertaken against government forces and 21 against Houthi forces, perpetuating the gradual attention shift from the state to the Houthis. While the group maintained its traditional focus on the lightly-populated east, undertaking 10 attacks in al-Bayda, eight attacks in Hadramout, and six in Shabwa, it also penetrated the densely populated, more urban, interior governorates, such as Dhamar (one operation) Ibb (two), and Sana’a (10). Fixed positions or structures were targeted the most (government positions six times and Houthi positions 10 times), through a combination of VBIEDs and mortar fire followed by a rapid assault by small teams of militants. This “hit-and-assault” tactic was balanced well with the use of 13 reported ambushes and six roadside IEDs on nine government vehicles, three Houthi vehicles, five government units, and five Houthi units. The types of targets of attack are constant with the previous months, but differ in terms of a higher rate (there was more than one attack on the same day 11 times, compared to eight times in August and three in September).

AQAP continued its aggressive approach to fixed positions in November by hitting a total of 24. However, 18 of these assaults were upon Houthi positions, part of a further reorientation that saw 36 of 52 overall attacks target Houthis (18 positions, 10 vehicles, four units, and four individuals). In terms of AQAP order of priorities, the Houthis were number one by November. This pattern is reinforced by the fact that 25 of the total attacks occurred in al-Bayda governorate, 24 of which were in Rada’a district. Rada’a — historically an al-Qaeda-dominated area — is strategically vital for both sides as it sits in between the AQAP-influenced east and Houthi-dominated west. AQAP continued to penetrate Houthi-held territory in Sana’a (five attacks), Dhamar (five), and Ibb (two). Continuing past efforts, AQAP persisted in consolidating control over the eastern governorates, striking predominantly government forces in Hadramout (eight times), Abyan (five), and Shabwa (two). AQAP was beginning to develop an assassination campaign where operatives were able to infiltrate highly-urbanized centers such as Sana’a and Dhamar and assassinate specific Houthi or Houthi-linked figures. At the same, AQAP’s more mobile fighting units were clearing already-held territory of Houthi and government personnel carriers and fixed, isolated bases.

The sharp emphasis on Rada’a in November was partly due to the Houthi advance that precipitated the withdrawal of AQAP forces from the area. This meant that AQAP could re-concentrate its resources into carrying out its war of attrition strategy from December into 2015. AQAP operations in December were even more sectarian than November, with 34 of 46 attacks against Houthi targets (19 fixed positions, eight individuals, six vehicles, and one unit). In line with popular jihadist belief that Iran and the U.S. are working together to back the Houthis to oppress Sunni Muslims, AQAP operatives attacked the Iranian ambassador to Yemen with an IED, and the al-‘Anad U.S. Air Force base in Lahj governorate. Notably, the group managed to conduct assassination attempts, ambushes, and surprise IED attacks against both Houthi and government forces in areas with little previous al-Qaeda influence, such as in Sana’a (seven attacks), the governorates of Ibb (four), Dhamar (three), al-Hudaydah (two) in the west, ‘Amran (one) in the north, and Ta’iz (one) in the southwest. AQAP also displayed remarkable operational mobility in attacking different parts of the country simultaneously. For example, on December 2, AQAP conducted operations in five different governorates (al-Hudaydah, ‘Amran, Abyan, Sana’a, and Ibb). Despite the setback in Rada’a in November, AQAP showed remarkable resilience in the number of attacks it managed to conduct, the variety of its targets (21 total fixed positions, 13 individuals, and 12 vehicles), and its geographic scope (11 different governorates).

In nominal terms, AQAP was most aggressive in the month of January — recording a total of 62 attacks. Forty-one of these operations were against Houthi targets, 20 of which on vehicles and personnel carriers, eight on fixed positions, seven on stand-alone units, five on individuals, and one on a religious structure — the Zayd Muslah cultural center in Sa’ada. Not surprisingly, 38 of these attacks against Houthis occurred in the Houthi-controlled or influenced governorates of al-Bayda, al-Dali’, Ibb, Sana’a, Dhamar, Sa’ada, and Marib. In AQAP’s heartland of Hadramout, Shabwa, and Abyan, there were a total of 18 attacks, in which 15 of them were against government forces (nine vehicles, five individuals, five units, and four fixed positions). Whereas in December, 75 percent of attacks on Houthis occurred in Houthi controlled or influenced territory, that figure rose to 93 percent in January. Similarly, the percentage of government-targeted operations in the area in which AQAP is strong rose from 50 percent in December to 83 percent in January. The pattern of clearing already-held territory of government forces — arguably weaker and more fractured than the Houthi militia — while penetrating and attacking Houthis on their own turf and in disputed zones has taken greater significance as the country inched closer to the current crisis.

A number of emerging patterns in AQAP operational activity in the last six months were reversed in February. First, out of 39 total attacks, 26 were against government targets (11 vehicles, seven individuals, five units, and three fixed positions) and only 15 were aimed at Houthi targets (five units, four individuals, four fixed positions, and two vehicles). Second, the most widely used type of attack was assassinations, with a total of 13 (only two of which did not succeed). And third, the share of government-targeted operations occurring in AQAP-influenced governorates dropped back to half, with the remaining 13 taking place in Houthi-held governorates.

Whereas February saw reversals of emerging patterns in AQAP operations, March represented a return to them. The group again emphasized attacks on Houthis, hitting the militia 44 times (18 fixed positions, 12 vehicles, eight units, and six individuals) out of 53 total operations in March. All Houthi-targeted operations occurred in al-Bayda, Lahj, Sana’a, and Ta’iz. All 41 operations in al-Bayda targeted Houthi forces, drawing parallels to November. All five operations in AQAP’s heartland (four in Hadramout and one in Abyan) were against government forces, in line with the pattern that emerged especially in January. In terms of tactics, there was constancy in that offensive assaults (14), infiltration (three), shelling (three), and VBIEDs (two) were often deployed against fixed positions; roadside IEDs (13) against vehicles; and sniper fire (six) and ambushes (two) against specific individual targets and units.

In April so far, there have been a total of 27 operations — 12 in al-Bayda, five in Lahj, four in Shabwa, three in Abyan, two in Taiz, and one in al-Hudaydah. AQAP continues to predominantly attack Houthi forces (eight vehicles, eight positions, seven units, and one individual) over government forces (one unit and one individual). This is the highest yet proportion in a pattern that has continued to increasingly prioritize Houthi over government targets, and strongly suggests that not much has changed in AQAP’s strategy since the Saudi campaign started on March 25. Despite only 27 operations, AQAP has deployed nine different types of attack: seven roadside IEDs, three VBIEDs, three bombings, three snipers, three IEDs, two assassinations, two ambushes, two offensive assaults, and one infiltration.

Drawing Conclusions from the Stats

By publicizing its attacks in the way that it has, AQAP has given analysts a data set with which to draw important conclusions. To be sure, the numbers remain too small and AQAP remains too dynamic to be able to predict its future orientation based solely on this statistical analysis. But there are discernible trends that should inform our understanding of the group, and its tactics offer potential clues about its evolving strategy. One of the more important trends is the growing number of attacks against Houthis. At the same time, AQAP has increasingly adopted a bifurcated strategy of targeting Houthis in Houthi areas and government targets in areas where AQAP is strongest. Finally, during the period analyzed, the group has demonstrated a consistent diversity of styles of attack, deploying at least eight different attack types in each month. Such a wide variety indicates AQAP’s ability to adapt to the enemy’s form via local support networks regardless of geographic location.

By their nature, organizations like AQAP are unpredictable. Strategic surprise is an imperative for terrorist groups. But analyzing data from a group’s recent operations provides telling indicators of its strategic orientation. And in the case of AQAP, such an analysis makes one particular point clear: While airstrikes in Yemen are changing the nature of AQAP’s battlespace, they are not yet fundamentally changing the strategic calculus behind a campaign that took shape and evolved well before the Saudi-led intervention.


*Unfortunately, there is a two week gap at the end of December 2014 and early January 2015 as a consequence of me being on vacation and the account being taken down in the intervening time frame, and therefore the data was lost. Additionally, as a consequence of repeated takedowns of its official account on Twitter, we retain a full archive of all claims on our computers.


Aaron Y. Zelin is the Richard Borow Fellow at the Washington Institute for Near East Policy and Rena and Sami David Fellow at the International Centre for the Study of Radicalisation and Political Violence. He is also a PhD candidate at King’s College London and founder of Jihadology.net


Patrick Hoover is a former research intern at the Washington Institute for Near East Policy and soon-to-be research assistant at a start-up CVE think tank in Washington D.C.

We have retired our comments section, but if you want to talk to other members of the natsec community about War on the Rocks articles, the War Hall is the place for you. Check out our membership at warontherocks.com/subscribe!

One thought on “What AQAP’s Operations Reveal about Its Strategy in Yemen

  1. Are the attacks against the government in traditional AQAP areas a deliberate preference to fight the government, or is there merely a absence of Houthi militia? Do the Houthi have any real presence in Hadramout, for example, or is the government the only real target there?

    Is the bifurcated strategy really much of a strategy, or merely the manifestation of a pattern of attacks of opportunity against whoever is strongest?