A Crude Threat:The Limits of An Iranian Missile Campaign Against Saudi Arabian Oil

A Crude Threat :The Limits of an Iranian Missile Campaign
against Saudi Arabian Oil
The MIT Faculty has made this article openly available. Please share
how this access benefits you. Your story matters.
Citation Itzkowitz Shifrinson, Joshua R. and Miranda Priebe. “A Crude

Threat: The Limits of an Iranian Missile Campaign against Saudi
Arabian Oil.” International Security 36 (2011): 167-201.© 2011 by
the President and Fellows of Harvard College and the
Massachusetts Institute of Technology.
As Published http://dx.doi.org/10.1162/ISEC_a_00048
Publisher MIT Press
Version Final published version

Accessed Mon Sep 07 10:34:55 EDT 2015
Citable Link http://hdl.handle.net/1721.1/66242
Terms of Use Article is made available in accordance with the publisher's policy
and may be subject to US copyright law. Please refer to the
publisher's site for terms of use.
Detailed Terms
A Crude Threat
A Crude Threat Joshua R. Itzkowitz

Shifrinson and
The Limits of an Iranian Missile Miranda Priebe
Campaign against Saudi Arabian Oil
D
evelopments in Iran’s
ballistic missile program have made headlines over the last several years.1
In the United States, Senator Carl Levin calls Iran’s short- and medium-
range missile arsenal “the number one threat in the Middle East,” and
Undersecretary of Defense for Policy Michèle Flournoy argues, “The threat
from Iran’s short- and medium-range ballistic missiles is developing more rap-
idly than previously projected.”2 Israeli ofªcials similarly describe the Iranian
missile program as “a matter of grave concern.”3 States around the Persian
Gulf, meanwhile, worry that Iran’s missiles are meant to intimidate and ex-
tract political concessions from them, with Saudi Arabia’s King Abdullah not-
ing that the Iranians “launch missiles in the hope of putting fear in the people
and the world.”4
These worries are not simply abstract concerns. At a time when the United
States and its allies remain locked in a standoff with Iran over the latter’s nu-
clear program, states around the Persian Gulf fear that Iran would retaliate for
an attack on its nuclear program by striking regional oil installations and other
strategic targets.5 Iranian ofªcials have threatened to use ballistic missiles in
Joshua R. Itzkowitz Shifrinson and Miranda Priebe are doctoral candidates in the Political Science Depart-
ment at the Massachusetts Institute of Technology.
The authors thank Nathan Black, Eugene Gholz, Brendan Rittenhouse Green, Llewelyn Hughes,
Barry Posen, Andrew Radin, and the anonymous reviewers for their comments.
1. Ed Blanche, “Iran Stages Display of Missile Firepower,” Jane’s Missiles and Rockets, November
23, 2006; Alan Cowell and William J. Broad, “Iran Reports Missile Test, Drawing Rebuke,” New
York Times, July 10, 2008; and Dennis C. Blair, “Annual Threat Assessment of the U.S. Intelligence
Community for the Senate Select Committee on Intelligence,” 111th Cong., 2d sess., February 2,
2010.
2. Levin and Flournoy quoted in “The President’s Decision on Missile Defense in Europe,” hear-
ing before the Senate Armed Services Committee, 111th Cong., 1st sess., September 24, 2009.
3. Angela Balakrishnan, “Barack Obama Calls for Tougher Iran Sanctions after Missile Tests,”
Guardian, July 9, 2008.
4. “Cables Show Arab Leaders Fear a Nuclear Iran,” Der Spiegel, December 1, 2010, http://
www.spiegel.de/international/world/0,1518,731877,00.html/.
5. Kenneth Katzman, Iran: U.S. Concerns and Policy Responses, CRS Report for Congress (Washing-
ton, D.C.: Congressional Research Service, Library of Congress, October 26, 2010), Order Code RL
32048, pp. 51–52; Patrick Clawson and Michael Eisenstadt, “The Last Resort: Consequences of Pre-
ventive Military Action against Iran,” Policy Focus, No. 84 (Washington, D.C.: Washington Insti-
tute for Near East Policy, 2008), pp. 16–18; and Anthony H. Cordesman, “The United States, Israel,
the Arab States, and a Nuclear Iran: Part Five: Iranian Capabilities to Respond to a Preventive At-
tack” (Washington, D.C.: Center for Strategic and International Studies, October 7, 2008), pp. 91–
120.
International Security, Vol. 36, No. 1 (Summer 2011), pp. 167–201

© 2011 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
167
International Security 36:1 168
these attacks, with a senior military adviser to Ayatollah Ali Khamenei warn-
ing that Iran would ªre its missiles against oil reªneries and other critical in-
frastructure in the event of a U.S. or Israeli strike.6 At the same time, Iran’s
efforts to develop and test its ballistic missile arsenal have led states in the re-
gion to improve their defenses around critical infrastructure, including oil fa-
cilities, with U.S. assistance.7 In a telling comment, a representative of Saudi
Arabian King Abdullah told a senior U.S. ofªcial that he “worries more about
an Iranian missile launch against Saudi oil facilities than a terrorist attack . . .
because he can take preventive measures against terrorism but not against
Iranian missiles.”8
A successful Iranian missile attack on Persian Gulf oil installations would
have many of the same effects as a blockade of the Strait of Hormuz.9 By dis-
rupting oil production, a successful missile strike could reduce the supply of
oil on the world market and cause a spike in oil prices.10 Presumably, Iranian
retaliation would be designed to impose substantial economic costs on an
attacker.11
These concerns raise several questions. What are the capabilities of Iran’s
missiles? Likewise, what are the military vulnerabilities in oil networks?
In light of the above scenario and the alarming reports about the Iranian mis-
sile arsenal, does Iran really have the missile capabilities to disrupt oil pro-
duction? Is increased spending to harden oil infrastructure by Persian Gulf
states worthwhile, or might such funds be better spent elsewhere? Over-
all, what damage could Iran inºict with a missile campaign against Gulf
6. Michael Smith, “Iran Threatens Gulf Blitz If U.S. Hits Nuclear Plants,” Sunday Times, June 10,
2007.
7. Joby Warrick, “U.S. Steps Up Arms Sales to Persian Gulf Allies,” Washington Post, January 31,
2010; and David S. Cloud, “U.S. Set to Offer Huge Arms Deal to Saudi Arabia,” New York Times,
July 28, 2007.
8. John M. Guilfoil, “The Embassy Cables: Saudis Fear Iranian Missile Strike More Than Terror-
ism,” Blast, December 7, 2010, http://blastmagazine.com/2010/12/07/the-embassy-cables-saudis-
fear-iranian-missile-strike-more-than-terrorism/.
9. Caitlin Talmadge, “Closing Time: Assessing the Iranian Threat to the Strait of Hormuz,” Inter-
national Security, Vol. 33, No. 1 (Summer 2008), pp. 82–117; Anthony H. Cordesman, “Iran, Oil, and
the Strait of Hormuz,” CSIS Brief (Washington, D.C.: Center for Strategic and International
Studies, March 26, 2007), pp. 1–7; and Robert Strauss Center, “Strait of Hormuz: Assessing Threats
to Energy Security in the Persian Gulf,” University of Texas at Austin, August 2008, http://
hormuz.robertstrausscenter.org/threat.
10. In 2008, global oil exports totaled 60.8 million barrels per day (mbd), including 40.1 mbd of
crude oil and 20.7 mbd of reªned products. Approximately 20.3 mbd came from the Middle East,
which, given the location of oil ªelds, means the Persian Gulf region. See Organization of the
Petroleum Exporting Countries (OPEC), Annual Statistical Bulletin, 2008 (Vienna: OPEC, 2008),
pp. 34, 36, 38.
11. Katzman, Iran, pp. 51–52; Clawson and Eisenstadt, The Last Resort, pp. 16–18; and Angus
McDowall, “Iran Threatens to Cut Off Gulf Oil Exports If Nuclear Facilities Are Attacked,” Tele-
graph, June 28, 2008.
A Crude Threat 169
oil installations—speciªcally, in the scenario below, Saudi Arabian facilities—

and how would it go about launching such an attack?
Existing analyses of Iranian retaliatory options have not discussed the mis-
sile scenario. Instead, studies of Iranian options address Iran’s capacity to close
the Strait of Hormuz to oil tankers. Because the ªndings suggest that a block-
ade is unlikely to be wholly successful, prudent Iranian planners might con-
sider additional ways, including missile launches, attacks by special forces and
proxy groups, naval assaults, and conventional air campaigns, to disrupt the
ºow of Persian Gulf oil.12 Of these options, a missile campaign is among
the most plausible: aside from the concerns described above, U.S. intelligence
ofªcials have argued that Iran’s missile arsenal is “an integral part of its strat-
egy to deter and if necessary retaliate against forces in the region.”13 This
study’s analysis of the missile scenario builds upon past work on Iran’s mili-
tary capabilities and addresses the policy concerns identiªed above.
This article offers an initial answer to the questions highlighted above by ex-
amining whether Iran could use its ballistic missile arsenal to signiªcantly re-
duce Saudi Arabia’s oil production. Analyzing the threat to Saudi production
provides a “worst-case” analysis from the perspective of the United States and
its allies. Simply put, Saudi Arabia contains the largest conªrmed oil reserves
in the world, holds the greatest productive capacity of any state, and is the
world’s largest oil exporter.14 If all Saudi oil production were to cease, 9.2 mil-
lion barrels per day (mbd) of oil would be removed from world markets;
only the combined loss of Emirate, Iraqi, Kuwaiti, Omani, and Qatari oil pro-
duction would equal these ªgures.15 Moreover, Saudi Arabia produces less oil
than its maximum capacity. It has historically used its excess capacity to cush-
ion oil markets by increasing production in crises to ensure a stable world oil
supply. A successful Iranian missile strike could eliminate this cushion, mak-
ing prices more volatile until production came back online.16 All this means
12. Talmadge, “Closing Time,” especially p. 115.

13. Blair, “Annual Threat Assessment of the U.S. Intelligence Community for the Senate Select
Committee on Intelligence,” p. 21.
14. BP, Statistical Review of World Energy, 2010 (London: BP, June 2010), pp. 6–7; U.S. Energy Infor-
mation Administration (EIA), “Country Analysis Brief: Saudi Arabia,” Department of Energy, No-
vember 2009, pp. 2–3, 11; and U.S. Energy Information Administration, “Country Energy Proªles,”
Department of Energy, http://tonto.eia.doe.gov/country/index.cfm.
15. The combined production of Iraq, Kuwait, Oman, Qatar, and the United Arab Emirates was
approximately 9.2 mbd in 2008. See OPEC, Annual Statistical Bulletin, 2008, p. 21; and U.S. Energy
Information Administration, “Country Analysis Brief: Oman,” Department of Energy, August
2009, p. 2, http://www.eia.doe.gov/cabs/Oman/Oil.html.
16. Permanent Subcommittee on Investigations, Senate Committee on Homeland Security and
Governmental Affairs, “The Role of Market Speculation in Rising Oil and Gas Prices: A Need to
Put the Cop Back on the Beat,” S. Prt. 109-65, 109th Cong., 2d sess. (Washington, D.C.: U.S. Gov-
ernment Printing Ofªce, June 2006), p. 13; and Alexander Kwiatkowski, “Oil Swings May Widen
as Spare Capacity Shrinks: Energy Markets,” Bloomberg.com, June 28, 2010.
that Saudi Arabia represents the most lucrative “target” if Iran (or, indeed, any
state) is interested in reducing oil production to upset world energy markets.
Drawing from open sources, our analysis of Saudi Arabia’s oil infrastruc-
ture and Iran’s missile capabilities ªnds that Iran could not signiªcantly re-
duce Saudi exports using its existing missile stockpile. Further, redundancies
in Saudi infrastructure and limits on Iranian capabilities make some Saudi ex-
ports virtually impossible to disrupt. This does not mean that an Iranian mis-
sile campaign would be without cost: any missile campaign is almost certain to
cause a large spike in oil prices.17 Still, because we believe there would be
no real damage to Saudi oil installations or disruption in oil production, gov-
ernments could take steps, such as the release of strategic petroleum reserves,
to calm energy markets.18 In a military sense, the Iranian missile threat to
Saudi Arabian—and, by extension, Persian Gulf—oil is overstated.
Our conclusion suggests that concerns surrounding Iran’s capacity to retali-
ate for an attack on its nuclear program by launching missiles at Persian Gulf
oil installations are militarily unfounded. Although Iran has other ways to
penalize any nation that attacked its nuclear program, a missile campaign
against oil infrastructure should not be a signiªcant concern for policymakers.
Further, funds currently devoted to hardening the region’s oil infrastructure
and improving its missile defenses are unnecessary. If, however, regional ac-
tors remain concerned about the ballistic missile challenge, the funds devoted
to missile defense would be better spent adding backup, or “redundant,” facil-
ities to Gulf oil networks to mitigate the consequences of an attack, rather than
trying to stop damage from occurring. This conclusion would change only if
Iran begins to develop longer-range missiles that more effectively employed
Global Positioning System (GPS) guidance.
The remainder of this article proceeds in eight sections. The following sec-
tion speciªes the working assumptions of our analysis. The next section pro-
vides a detailed description of Saudi Arabia’s oil network and identiªes likely
Iranian targets. We then discuss Iran’s missile capabilities and Saudi Arabia’s
defensive assets. Subsequently, we analyze the requirements and effects of an
Iranian ballistic missile campaign. Next, the article considers potential Saudi
countermeasures and what our ªndings suggest for other forms of Iranian re-
17. Oil and gas prices often move in response to perceived vulnerabilities in energy supplies. The
economic effects of an attack may thus be out of proportion to actual supply disruptions. See Per-
manent Subcommittee on Investigations, “The Role of Market Speculation in Rising Oil and Gas
Prices,” pp. 1–4.
18. Olivier J. Blanchard and Jordi Gali, “The Macroeconomic Effects of Oil Shocks: Why Are the
2000s So Different from the 1970s?” NBER Working Paper, No. 13368 (Cambridge, Mass.: National
Bureau of Economic Research, September 2007).
A Crude Threat 171
taliation. The article concludes by discussing the implications of our analysis

for U.S. and allied policy in the Persian Gulf and understanding military vul-
nerabilities in oil networks writ large.
Setting and Assumptions
We do not consider an unprovoked Iranian attack on Saudi Arabian oil infra-

structure likely. Because an attack would likely invite a violent international
response, it is implausible that Iran would target these installations except
over vital national security issues. Given current security concerns, Iran’s use
of ballistic missiles against oil installations would most likely follow a U.S. or
an Israeli strike on Iran’s nuclear facilities. Some analysts might question
whether Iran would invite further punishment by retaliating against its neigh-
bors and a key U.S. interest. Nevertheless, American and Persian Gulf leaders
believe that Iran poses a real threat. Likewise, even if the scenario seems un-
likely, this analysis presents a worst-case scenario that allows us to explore
more general claims about Iran’s missile capabilities and Saudi Arabian vul-
nerabilities. To do so, we make several simplifying assumptions.
First, we assume that Iran’s goal would be to disrupt global oil supplies in
an effort to retaliate for an attack while depriving Saudi Arabia—whose sup-
port for efforts to curb the Iranian nuclear program are well documented—of
its principal source of revenue.19 In other words, the aim of the missile cam-
paign would be to prevent Saudi oil from reaching world markets. We there-
fore assume that the Iranian attack would employ all of Iran’s missile assets.
Second, we consider the maximum damage Iran could cause given Saudi
Arabia’s independent capability to defend its oil network. We therefore as-
sume that the United States does not become involved in the conºict. This as-
sumption helps establish the absolute magnitude of the Iranian threat to
Persian Gulf oil. If Saudi Arabia can defend its oil installations using its own
forces, then analysts can challenge the notion of an Iranian threat to regional
security. Conversely, if Iran can effectively shut down the Saudi oil network,
then there is greater justiªcation for efforts to improve Saudi Arabian defenses.
Third, we premise our analysis on a near-term clash between Iran and Saudi
Arabia in which both sides ªght with the military capabilities already in their
arsenals.
Fourth, we assume that Iran enjoys access to perfect information regarding
the location of Saudi oil facilities. Even if perfect information is unavailable,
19. Ian Black and Simon Tisdall, “Saudi Arabia Urges U.S. Attack on Iran to Stop Nuclear
Programme,” Guardian, November 28, 2010.
Iran could likely obtain very good information using open source services
(e.g., Google Earth) and intelligence collected by Iranian agents.
Fifth, we assume that all of Iran’s missiles ªre and detonate as intended,
without any “duds.”20 Although the fourth and ªfth assumptions are unlikely
in practice, they maximize Iran’s chance of success in accordance with the
worst-case-scenario nature of this exercise.
Finally, we consider only an Iranian ballistic missile attack. In practice, Iran
could conduct special operations, air, and naval attacks alongside a missile
strike.21 It might also divert assets to attack shipping in the Persian Gulf or to
blockade the Strait of Hormuz.22 We focus exclusively on a ballistic missile
campaign for analytic clarity, though we discuss the implications of this analy-
sis for other forms of attack later in the article.
Saudi Arabia’s Oil Infrastructure and Its Vulnerabilities
Saudi Arabia has the largest proven crude oil reserves in the world, with ap-
proximately 20 percent of the world total.23 These reserves are distributed
among eighty-ªve oil ªelds containing more than 1,000 producing wells,
though most production comes from six to eight ªelds.24 Saudi oil production
is managed by Saudi Aramco, the state-owned oil company. Crude oil produc-
tion averaged approximately 9.2 mbd in 2008 out of a potential capacity of ap-
proximately 11.8 mbd.25 This represents nearly 13 percent of oil produced
20. During the Persian Gulf War, nearly 10 percent of Iraqi Scud missiles ªred at Israel failed to ex-
plode. See George N. Lewis, Steve Fetter, and Lisbeth Gronlund, “Casualties and Damage from
Scud Attacks in the 1991 Gulf War,” DACS Working Paper (Cambridge, Mass.: Defense and Arms
Control Studies Program, Center for International Studies, Massachusetts Institute of Technology,
March 1993), pp. 29–30.
21. For example, the Iranian navy attacked Iraqi oil ports during the Iran-Iraq War. See Dilip Hiro,
The Longest War: Iran-Iraq Military Conºict (New York: Routledge, 1991), p. 47.
22. Talmadge, “Closing Time.”
23. The size of Saudi and world oil reserves is a subject of debate, but the fact remains that Saudi
Arabia holds a signiªcant share of proven oil reserves worldwide. See BP, Statistical Review of
World Energy, 2010, pp. 6–7; and OPEC, Annual Statistical Bulletin, 2008, p. 18. For a critique of BP
and OPEC data, see Matthew R. Simmons, Twilight in the Desert: The Coming Saudi Oil Shock and the
World Economy (Hoboken, N.J.: Wiley and Sons, 2005).
24. Nawaf Obaid and Nathaniel Kern, “Saudi Arabia’s Upstream and Downstream Expansion
Plans for the Next Decade: A Saudi Perspective,” Working Paper (Washington, D.C.: Center for
Strategic and International Studies, October 2005), p. 5, http://csis.org/publication/saudi-arabias-
upstream-and-downstream-expansion-plans-next-decade-saudi-perspective; Anthony H. Cordes-
man and Nawaf Obaid, National Security in Saudi Arabia: Threats, Responses, and Challenges
(Westport, Conn.: Praeger, 2005), p. 310; and EIA, “Country Analysis Brief: Saudi Arabia,” Novem-
ber 2009, pp. 2–3, 11. We thank one of the reviewers for pointing out that most production comes
from a small number of ªelds.
25. For production ªgures, see OPEC, Annual Statistical Bulletin, 2008, p. 21. For total capacity, see
Kate Dourian, “Saudi Arabia Starts Khursaniyah Oil Field Production,” Platts Oilgram Price Report,
September 4, 2008, p. 13. Slightly lower total production ªgures are reported in EIA, “Country
A Crude Threat 173
around the world daily.26 Saudi Arabia is also the world’s largest oil exporter
with total exports of 8.4 mbd.27 It is, however, a much more signiªcant player
in crude oil markets, with 18.2 percent of world crude exports (7.3 mbd),
than in reªned product markets (5.3 percent of world reªned exports with
1.1 mbd product exports).28
Once pumped from ªelds, oil travels to processing facilities throughout
Saudi Arabia via 15,000 kilometers (km) of pipelines and more than thirty
pumping stations. These “downstream” facilities prepare the oil for domestic
consumption or export.29
There are several basic steps in this process. Freshly pumped oil consists of
an unstable mixture of oil, water, gas, and sand that can damage industrial
equipment; non-oil elements must be removed before the oil can be further
processed. Oil is therefore pumped directly from the ªelds to one of sixty gas-
oil separation plants (GOSPs) where the elements are separated and the oil
prepared for further processing.30
After leaving a GOSP, the majority of Saudi oil moves to stabilization plants
for further treatment. Except for the approximately 2.6 mbd capacity found in
the Central Arabian, Safaniya, Shaybah, and Zuluf ªelds, all Saudi oil is con-
sidered “sour”: that is, it contains signiªcant levels of hydrogen sulªde.31
Analysis Brief: Saudi Arabia,” November 2009, p. 2; and Steven Mufson, “Saudi Arabia Works the
Vast Desert to Pump Out More High-Quality Oil,” Washington Post, November 17, 2007.
26. Total world production of 72.9 mbd is the average of 2008 production ªgures provided by
OPEC and DOE/EIA. See OPEC, Annual Statistical Bulletin, 2008, p. 22; and U.S. Energy Informa-
tion Administration, “World Crude Oil Production, 1960–2009,” in Annual Energy Review, 2009
(Washington, D.C.: EIA, Department of Energy, August 2010), p. 315, http://www.eia.doe.gov/
aer/inter.htm.
27. This represents 14.8 percent of world exports. World export totals are the sum of crude and
reªned exports reported in OPEC, Annual Statistical Bulletin, 2008, pp. 34, 36.
28. Percentages are based on global crude exports of 40.1 mbd and reªned exports of 20.7 mbd.
See ibid., pp. 21–37.
29. EIA reports 9,000 miles of pipeline. See U.S. Energy Information Administration, “Country
Analysis Brief: Saudi Arabia,” Department of Energy, February 2007, p. 11. Saudi Arabia’s major
domestic pipeline is supported by eleven pump stations. Additionally, a 1998 U.S. military map of
Saudi Arabian infrastructure shows more than twenty pump stations connecting ªelds and local
facilities; John M. Collins, Military Geography for Professionals and the Public (Washington, D.C.: Na-
tional Defense University Press, July 1998), p. 163.
30. Simmons, Twilight in the Desert, p. 104; Sunil L. Kokal and Abdullah Al-Ghamdi, “Performance
Appraisals of Gas/Oil Separation Plants,” SPE Production & Operations, Vol. 23, No. 2 (May 2008),
pp. 287–298; and Ismail I. Nawwab, Peter C. Speers, and Paul F. Hoye, eds., Aramco and Its World:
Arabia and the Middle East, 2d ed. (Dhahran, Saudi Arabia: Aramco, 1981), p. 215.
31. There is some confusion surrounding the term “sour” oil. We follow Aramco in using it to re-
fer to oil that contains hydrogen sulªde. See Arthur P. Clark, Muhammad A. Tahlawi, William
Facey, and Thomas A. Pledge, A Land Transformed: The Arabian Peninsula, Saudi Arabia, and Saudi
Aramco (Dhahran, Saudi Arabia: Saudi Arabian Oil Co., 2006), p. 274; and Saudi Aramco, “Abqaiq
Plants,” http://www.saudiaramco.com/irj/portal/anonymous?favlnk⫽/SaudiAramcoPublic/
docs/Our⫹Business/Oil⫹Operations/Production⫹Facilities/Abqaiq⫹Plants&ln⫽en. The capac-
ities of Zuluf and Safaniya are based on 660,000 mpd capacity in Zuluf and 1.2 mbd capacity in
Hydrogen sulªde makes sour oil dangerous to transport via tanker because it
is poisonous in its gaseous form and highly corrosive. The sour oil must there-
fore be “sweetened” by removing the hydrogen sulªde before it can be
shipped to world markets.32 This process—referred to as “stabilization”—
occurs at one of ªve facilities in Saudi Arabia.33 Abqaiq is by far the most im-
portant of these facilities, as it processes two-thirds of all Saudi oil (6.1 mbd)
and has a potential capacity of 13 mbd.34 Total capacity of the smaller plants is
approximately 3.0 mbd.35
Safaniya. See Ivan Sandrea and Rafael Sandrea, “Growth Expected in Global Offshore Crude Oil
Supply,” Oil and Gas Journal, March 12, 2007, p. 34. Shaybah capacity (0.5 mbd) is from Amy Myers
Jaffe and Jareer Elass, “Case Study on Saudi Aramco,” presentation at the James A. Baker III Insti-
tute for Public Policy, March 1–2, 2007, p. 6, http://www.bakerinstitute.org/programs/energy-
forum/publications/docs/NOCs/Presentations/Hou-Jaffe-SaudiAramco.pdf. The Central Ara-
bian ªelds produce Arabian Super-Light oil, production of which averaged 0.250 mbd in 2004. See
Anthony H. Cordesman and Nawaf Obaid, “Saudi Petroleum Security: Challenges and Re-
sponses,” Working Paper (Washington, D.C.: Center for Strategic and International Studies, No-
vember 30, 2004), p. 9, http://csis.org/publication/saudi-petroleum-security; and Adurty V.C.
Rao, O.M. Halawani, I.A. Mulhim, and B.M. Zayed, “Artiªcial Lift System Operation at Saudi
Aramco Central Arabia Fields,” presentation at the Seventeenth World Petroleum Congress, Rio
de Janeiro, Brazil, September 1–5, 2002.
32. Nawwab, Speers, and Hoye, Aramco and Its World, p. 216.
33. The ªve stabilization facilities are located at Abqaiq, Juaymah, Jubail, Qatif, and Ras Tanura.
Three facilities—Abqaiq, Ras Tanura, and Ras al-Juaymah—were reported in 2005. See Cordesman
and Obaid, National Security in Saudi Arabia, p. 314. The 2007 EIA Country Analysis Brief included
stabilization facilities for the planned Qatif Producing Plants program. See U.S. Energy Informa-
tion Administration, “Country Analysis Brief: Saudi Arabia,” Department of Energy, February
2007, p. 5. Aramco recently announced that the new Khursaniyah program includes facilities near
Jubail to “process and stabilize” 500,000 bpd crude oil from the Abu Hadriya, Fadhili,
and Khursaniyah ªelds. See Saudi Aramco, “Khursaniyah Program,” http://www.saudiaramco
.com/irj/portal/anonymous?favlnk⫽/SaudiAramcoPublic/docs/Our⫹Business/Projects/
Khursaniyah&ln⫽en; and “Saudi Aramco Khursaniyah Oil and Gas and Hawiyah NGL Recovery
Programmes, Saudi Arabia,” http://www.hydrocarbons-technology.com/projects/saudi-aramco.
Aramco previously noted the presence of stabilization facilities at Abqaiq, Juaymah, Ras Tanura,
and Abu Hadriya. See Nawwab, Speers, and Hoye, Aramco and Its World, p. 216. No other source
mentions an active stabilization plant at Abu Hadriya. EIA reports that the ªeld was shut down in
the 1980s. See EIA, “Country Analysis Brief: Saudi Arabia,” February 2007, p. 7.
34. On the two-thirds ªgure, see Cordesman and Obaid, National Security in Saudi Arabia, p. 315.
Daily throughput was calculated by multiplying all Saudi production by two-thirds. The two-
thirds ªgure is supported by reports that Abqaiq is the main processing facility for Arabian Light
and Extra Light oil blends, which together make up roughly 71 percent of Saudi productive capac-
ity. See Aramco, “Abqaiq Plants”; and Khalid R. al-Rodhan and Nawaf Obaid, “Saudi Arabia’s
Sustainable Capacity and Security Issues,” Working Paper (Washington, D.C.: Center for Strategic
and Internationl Studies, May 2005), p. 10, http://csis.org/publication/saudi-arabias-sustainable-
capacity-and-security-issues. Aramco reports that Abqaiq processes nearly 70 percent of all Saudi
production. See Clark et al., A Land Transformed, p. 274. The New York Times is one of the few
sources on Abqaiq’s actual capacity (13 mbd); Saudi Aramco simply reports that Abqaiq’s capacity
is “more than 7 million barrels a day.” See Hassan M. Fattah, “Suicide Bombers Fail to Enter Saudi
Oil Plant,” New York Times, February 25, 2006.
35. The Ras Tanura stabilization plant has a capacity of 0.96 mbd; Qatif and Jubail have 0.8 and
0.5 mbd capacity, respectively; and we estimate Juaymah at 0.75 mbd. For the Ras Tanura ªgure,
see Saudi Aramco, “Ras Tanura Reªnery,” http://www.saudiaramco.com/irj/portal/anonymous
?favlnk⫽/SaudiAramcoPublic/docs/Our⫹Business/Reªning⫹%26⫹Distribution/Reªning/Ras
⫹Tanura⫹Reªnery&ln⫽en; for the Jubail ªgure, see Saudi Aramco, “Khursaniyah Program.” The
A Crude Threat 175
From a stabilization plant, crude oil is pumped either directly to a port

for shipment abroad or to a reªnery for processing into commercial prod-
ucts (e.g., gasoline). In the latter case, crude oil is moved to one of seven re-
ªneries.36 After processing, the reªned product intended for export is pumped
to Saudi ports for loading onto oil tankers.
Saudi Arabia’s ports can export more than 15.5 mbd of combined crude and
reªned product.37 Its three major oil ports are located at Ras Tanura and Ras al-
Juaymah on the Persian Gulf and Yanbu on the Red Sea.38 Additional capacity
is found at a series of smaller ports at Jeddah, Jizan, and Rabigh on the Red
Sea, and Jubail, Ras al-Khafji, and Zuluf on the Persian Gulf.39 As a rule of
thumb, Ras Tanura and Juaymah handle approximately 75 percent (6.3 mbd)
of all Saudi oil exports; most of the remaining 25 percent ships from Yanbu.40
Moving oil to the Red Sea ports depends more on pumping stations than
Qatif stabilization plant became operational in 2004 and is intended to stabilize oil from the Qatif
and Abu Safa ªelds. These ªelds produced 0.8 mbd in 2004, meaning the plant is at least that large.
See EIA, “Country Analysis Brief: Saudi Arabia,” February 2007, p. 5; and Cordesman and Obaid,
National Security in Saudi Arabia, p. 313. We estimate Ras al-Juaymah stabilization capacity as the
average capacity from Ras al-Juaymah, Qatif, and Jubail.
36. EIA, “Country Analysis Brief: Saudi Arabia,” November 2009, p. 11.
37. This includes approximately 0.25 mbd exported via pipeline to Bahrain. See Geoff King, “Bah-
rain, Saudi Arabia in 350,000 b/d Pipeline Talks,” Platts Oilgram News, October 30, 2009; and Saudi
Economic Survey, “Bahrain to Import 25% More Crude from Saudi Arabia,” October 2, 2002.
38. Ras Tanura has a capacity of at least 6 mbd. See EIA, “Country Analysis Brief: Saudi Arabia,”
November 2009, pp. 7, 10; and Amy Myers Jaffe and Jareer Elass, “Saudi Aramco: National Flag-
ship with Global Responsibilities” (Houston, Tex.: James A. Baker III Institute for Public Policy,
Rice University, March 2007), p. 86. Ras al-Juaymah has a reported capacity of 3–3.5 mbd. See EIA,
“Country Analysis Brief: Saudi Arabia,” November 2009, p. 7; Nawaf Obaid, “Saudi Arabia’s Stra-
tegic Energy Initiative: Safeguarding against Supply Disruptions,” presentation at the Center for
Strategic and International Studies, Washington, D.C., November 9, 2006, http://www.csis.org/
ªles/media/csis/events/061109_omsg_presentation1.pdf, p. 13; and EIA, email correspondence
with authors, May 9, 2008.
39. Data on the capacity of the smaller ports are not available. For information on berths, see An-
thony H. Cordesman, Saudi Arabia: Guarding the Desert Kingdom (Boulder, Colo.: Westview, 1997),
p. 85; Jaffe and Elass, “Saudi Aramco,” p. 86; and M. Webster Ewell, Dagobert Brito, and John
Noer, “An Alternative Pipeline Strategy in the Persian Gulf,” Working Paper (Houston, Tex.: James
A. Baker III Institute for Public Policy, Rice University, March 2007), p. 7, http://www.rice.edu/
energy/publications/docs/TrendsinMiddleEast_AlternativePipelineStrategy.pdf. Jubail and Jed-
dah are conªgured for reªned product and petrochemicals and may be unable to export crude oil.
EIA, email correspondence with authors; National Geospatial Intelligence Agency (NGIA), Sailing
Directions: Red Sea and Persian Gulf, 15th ed. (Bethesda, Md.: U.S. Government, 2009), pp. 115–118,
349–351; and Saudi Aramco, “Jiddah Reªnery Marine,” http://www.saudiaramco .com/irj/
portal/anonymous?favlnk⫽%2FSaudiAramcoPublic%2Fdocs%2FOur⫹Business%2FReªning⫹
%26⫹Distribution%2FPorts⫹%26⫹Terminals%2FJiddah⫹Reªnery⫹Marine&ln⫽en. We exclude
further discussion of Jizan from our analysis because of its extremely small capacity. See Saudi
Aramco, “Jizan Marine,” http://www.saudiaramco.com/irj/portal/anonymous?favlnk⫽
%2FSaudiAramcoPublic%2Fdocs%2FOur⫹Business%2FReªning⫹%26⫹Distribution%2FPorts⫹
%26⫹Terminals%2FJizan⫹Marine&ln⫽en; and NGIA, Sailing Directions, pp. 134–135. Finally, al-
though it is unclear whether the Zuluf oil terminal is in use, we follow Jaffe and Elaas in consider-
ing it operational. See Jaffe and Elass, “Saudi Aramco,” p. 86; and NGIA, Sailing Directions, p. 355.
40. EIA, “Country Analysis Brief: Saudi Arabia,” February 2007, pp. 19–20; and EIA, “Country
Analysis Brief: Saudi Arabia,” November 2009, p. 7.
Table 1. Overview of Saudi Arabian Production and Facilities
Quantity
Category (in millions of barrels per day)
Productive capacity ⬎ 11.8

requiring stabilization ⬎ 9.2
not requiring stabilization ⬎ 2.6
Stabilization capacity ⬎ 16.0
Abqaiq ⬎ 13.0
Other facilities ⬎ 3.0
Amount produced ⬎ 9.2
stabilized at Abqaiq ⬎ 6.1 (est.)
stabilized elsewhere ⬎ 1.25 (est.)
not requiring stabilization ⬎ 1.85 (2003 est.)
Amount exported ⬎ 8.4
as crude oil ⬎ 7.3
as reªned product ⬎ 1.1
Export capacity ⬎ 15.5
Ras Tanura and Juaymah ⬎ 9.0–9.5
Yanbu ⬎ 6.5
Other unknown
does transporting oil to the Gulf ports. Because Ras Tanura is less than 100 ki-
lometers downhill from Abqaiq, oil can likely ºow from Abqaiq to Ras Tanura
largely by force of gravity. At least three pump stations line the route, but these
may not be necessary for oil to reach the terminals.41 When oil is moved uphill,
however, pump stations are necessary to overcome the force of gravity.42
Given that Yanbu and the other Red Sea facilities are more than 1,200 km to the
east, mostly uphill, from Abqaiq, eleven pumping stations are used to move oil
through the 1,400 km, 5 mbd “Petroline” connecting Abqaiq to the Red Sea
oil ports.43 Table 1 summarizes the preceding discussion of Saudi production
and capacity.
We assume that an Iranian attack on Saudi Arabia’s oil infrastructure would
41. Collins, Military Geography for Professionals and the Public, p. 163. A U.S. Army manual notes
that pump stations are necessary only when gravity is insufªcient to maintain oil ºow or when
one desires a higher ºow rate than is achievable by gravity, or both. See U.S. Army Field Manual,
“Chapter 4–Design,” in Military Petroleum Pipeline Systems, FM No. 5-482 (Washington, D.C.: De-
partment of the Army, August 1994), http://www.globalsecurity.org/military/library/policy/
army/fm/5-482/index.html. For topographical information and distances, see Oxford, Atlas of the
World, 10th ed. (New York: Oxford University Press, 2002), pp. 96–99.
42. For instance, Iranian attacks during the Iran-Iraq War on Iraqi pump stations reduced
throughput on the Iraq-Turkey pipeline running uphill from the Kirkuk area. See “Iran and Iraq: A
New Front in a Slow War?” Economist, January 3, 1981, p. 28; Roger Vielvoye, “Iran/Iraq War
Takes Toll of Oil Facilities,” Oil and Gas Journal, October 6, 1980, p. 48; and “Iraqi Crude Exports
May Rise Further,” Oil and Gas Journal, December 8, 1980, p. 54.
43. EIA, “Country Analysis Brief: Saudi Arabia,” February 2007, p. 11.
A Crude Threat 177
try to cause the maximum amount of damage with the least possible expendi-
ture of force. To select targets, we hypothesize that Iran would consider the
quantity of oil ºowing through each facility, redundancy of each facility (i.e.,
how readily a similar facility could replace its functions), size (i.e., concentra-
tion) of the target, and speed with which the targeted facility could be
repaired.
Some facilities are unlikely targets because they are dispersed or redundant.
For example, Iran is unlikely to attack Saudi oil ªelds. To stop production at an
oil ªeld, Iran would have to destroy the wells. This would require the destruc-
tion of many small targets spread over a large area; to halve production, for
example, Iran would have to eliminate more than 500 wells spread across sev-
eral thousand square miles.44 GOSPs make poor targets for similar reasons:
although not as numerous as wells, more than sixty such facilities are spread
throughout the country.45
Nor is Iran likely to target pipelines, given their small size and ease of repair.
Not only are there more than 15,000 km of pipeline in the country, but Aramco
has taken steps to minimize the effects of pipeline damage. First, cameras
and monitoring systems help to identify damage and expedite repairs. Sec-
ond, pipelines have shutoff valves to limit oil losses from a rupture.46 Third,
Aramco pre-positions replacement parts throughout the country and can re-
portedly repair damage to pipelines within thirty-six hours.47
Finally, we do not believe that Iran would focus on Saudi Arabian oil
reªneries. As noted above, Saudi Arabia’s reªned products constitute a small
percentage of Saudi exports and are comparatively less important as a percent-
age of global oil supplies compared with crude oil. Although we cannot totally
discount an Iranian attack on Saudi reªneries, the disparity in crude and re-
ªned production suggests that Iran’s efforts would be best served elsewhere.
We conclude that Iran would target Saudi Arabian stabilization facilities.
Five factors underlie this ªnding. First, destruction of the stabilization plants—
with Abqaiq a particularly lucrative target—would prevent Saudi Arabia from
transforming its sour crude into a product safe for export. Second, Saudi stabi-
lization facilities have been targeted previously when a Saudi terrorist cell un-
44. The Ghawar Field alone covers 2,600 square miles. For the size of the oil ªeld, see John C.K.
Daly, “Saudi Oil Facilities: Al-Qaeda’s Next Target?” Terrorism Monitor, Vol. 4, No. 4 (February
2006), p. 6.
45. GOSP capacity can vary, but even the newest facilities have a capacity of only 300,000 barrels
per day. See Cordesman and Obaid, “Saudi Petroleum Security,” pp. 10–11.
46. J.J. Dempsey, A.H. Al-Gouhi, and F.T. Connor, “A Computer Simulation of the Saudi Aramco
East/West Crude Pipeline,” Pipeline Simulation Interest Group, No. 9202 (1992), http://
www.psig.org/papers/1990/9202.pdf, pp. 1–3; and Cordesman and Obaid, National Security in
Saudi Arabia, p. 323.
47. Cordesman and Obaid, National Security in Saudi Arabia, p. 323.
successfully attacked Abqaiq with truck bombs.48 Third, although the facilities
are large (Abqaiq alone measures almost 3 km2), the stabilization process oc-
curs in towers concentrated in speciªc parts of each facility.49 Fourth, each sta-
bilization facility is within 300 km of Iran and thus within range of most
Iranian missiles.50 Finally, some of the stabilization towers were speciªcally
designed for Saudi facilities, meaning they would take a signiªcant amount of
time to replace.51
As a next-best option, Iran might try to prevent oil from reaching the market
by attacking the Saudi export system. Compared to the stabilization plants, the
export system is a second-best option because of the small size of the targets
and excess capacity. Nevertheless, we include the scenario because of the prox-
imity of the Gulf ports to Iran. There is also a long history of warring states tar-
geting oil export facilities. During World War II, for instance, the Allies
targeted rail-lines and ports transporting Romanian oil to Germany.52 Like-
wise, the Iran-Iraq War saw each country try to impede the other’s oil exports
by attacking export facilities. Iran, for instance, attacked Iraqi oil terminals off
the Fao Peninsula, and the Iraqi air force launched a bombing campaign
against Iran’s Kharg Island terminal.53
The primary target of such an attack would be the Gulf ports through which
most Saudi oil is shipped. Attacking the Gulf ports alone, however, would still
enable Saudi Arabia to export at least 5 mbd (59 percent of current exports)
through its Red Sea facilities.54 Therefore, an attack on the export system might
also target the Red Sea ports and pump stations along the Petroline.55
48. Hassan M. Fattah, “Attack on Saudi Oil Facility Thwarted,” New York Times, February 24, 2006;
and Gal Luft, “An Energy Pearl Harbor? A Near Miss in Saudi Arabia Hints at Future Shocks,”
Washington Post, March 5, 2006.
49. Robert Baer, Sleeping with the Devil: How Washington Sold Our Soul for Saudi Crude (New York:
Crown, 2003), pp. xxi–xxiii; and Khalid al-Rodhan, “The Impact of the Abqaiq Attack on Saudi En-
ergy Security,” Working Paper (Washington, D.C.: Center for Strategic and International Studies,
February 2006), p. 3, http://csis.org/publication/impact-abqaiq-attack-saudi-energy-security.
50. National Geographic, Atlas of the World, 7th ed. (Washington, D.C.: National Geographic Soci-
ety, 2000), pp. 95–96; and Google Earth, January 2010.
51. Oil Fields as Military Objectives: A Feasibility Study, CRS Report for Congress (Washington, D.C.:
Congressional Research Service, Library of Congress, U.S. Government Printing Ofªce, 1975), pp.
1–39.
52. Ronald C. Cooke and Roy Conyers Nesbit, Target, Hitler’s Oil: Allied Attacks on German Oil Sup-
plies, 1939–1945 (London: William Kimber, 1985), pp. 70–77, 105–107.
53. Loren Jenkins, “Iraq and Iran Attack Each Other’s Oil Installations,” Washington Post, October
17, 1980; Hiro, The Longest War, p. 57; and Anthony H. Cordesman and Abraham Wagner, “Phase
Four: Stalemate and War of Attrition on Land,” in Cordesman and Wagner, eds., The Lessons of
Modern War, Vol. 2: The Iran-Iraq War (Washington, D.C.: Center for International and Strategic
Studies, 1990), chap. 7, pp. 5–7.
54. Exports of 5 mbd assume that Saudi Arabia has no prepositioned oil stocks on the Red Sea
coast and depends solely on the Petroline. If it has prepositioned stocks, then it could export sub-
stantially more, given that Yanbu alone has an export capacity of 6.5 mbd.
55. A 1991 study by Aramco engineers suggests that shutdown of the last pump station along the
Petroline would eliminate up to 56 percent of Petroline throughput. Robert Baer asserts that de-
A Crude Threat 179
Iranian and Saudi Arabian Forces
In the scenario we describe, Iran would attempt to destroy Saudi oil installa-
tions using its existing ballistic missile arsenal. Over the past two decades, Iran
has worked to improve its short- and intermediate-range ballistic missile
(SRBM and IRBM) capabilities. Since the end of the Iran-Iraq War, Iranian mis-
sile assets expanded from a handful of Scud-B missiles purchased from North
Korea (known locally as Shahab-1s) to a large collection of imported and do-
mestically produced missiles. Speciªc information on the Iranian stockpile is
scarce, but most Iranian missiles appear to be road-mobile; some are becoming
increasingly accurate.56 Utilizing technology from China, North Korea, and
Syria, Iran’s most advanced SRBMs may be able to obtain a 100-meter circular
error probable (CEP) with a system employing inertial guidance—possibly
with GPS updates—and limited terminal maneuvering.57 By way of contrast,
the Scud-B employed by Iraq in the Persian Gulf War had a CEP of more than
1,000 meters.58 There is no evidence, however, that Iran has taken the techno-
logical leap to successfully integrate GPS into a terminally guided missile able
to achieve accuracies on par with Western and other advanced systems.59
Table 2 presents a summary of the Iranian ballistic missile arsenal, excluding
programs under development, unconªrmed, or believed terminated.60
struction of the ªrst pump station on the Petroline (Pump Station 1) would stop the ºow of oil to
the Red Sea. Given the ambiguity of Baer’s report, we follow the Aramco study in assuming that
destruction of any one pump station would reduce rather than stop the ºow of oil. Baer, “The Fall
of the House of Saud,” Atlantic Monthly, May 2003, p. 54; and Dempsey, Al-Gouhi, and Connor,
“A Computer Simulation of the Saudi Aramco East/West Crude Pipeline,” p. 4.
56. Andrew Feickert, Iran’s Ballistic Missile Capabilities, CRS Report for Congress (Washington,
D.C.: Congressional Research Service, Library of Congress, August 23, 2004), Order Code RS21548,
pp. 1–3; Steven A. Hildreth, Iran’s Ballistic Missile Programs: An Overview, CRS Report for Congress
(Washington, D.C.: Congressional Research Service, Library of Congress, November 8, 2007), Or-
der Code RS22758; and Anthony H. Cordesman and Martin Kleiber, Iran’s Military Forces and
Warªghting Capabilities: The Threat in the Northern Gulf (Washington, D.C.: Center for Strategic and
International Studies Press, 2007), pp. 137–148.
57. CEP is deªned as the radius of the circle within which 50 percent of all missiles launched at a
target will impact. Duncan Lennox, ed., Jane’s Strategic Weapons Systems (hereafter JSWS), Vol. 50
(Alexandria, Va.: Jane’s Information Group, 2009), pp. 71–72; Jane’s Information Group, “Iran Tests
Fateh 110A Ballistic Missile,” Jane’s Missiles and Rockets, September 23, 2002; Jane’s Information
Group, “Iran Tests Solid-Propellant Ballistic Missile,” Jane’s Missiles and Rockets, June 15, 2001; and
“Fateh-110/NP-110/Mushak,” GlobalSecurity.org, April 28, 2005, http://www.globalsecurity.org/
wmd/world/iran/mushak.htm.
58. Bernard Rostker, “Iraq’s Scud Ballistic Missiles,” Department of Defense, July 25, 2000, http://
www.globalsecurity.org/wmd/library/news/iraq/2000/scud_info/.
59. For comparison, the U.S. Army’s ATACMS Block 1A has a CEP of 10–50 meters. Russian and
Chinese designs that combine GPS with terminal guidance have CEPs of 10–30 meters. Lennox,
JSWS, pp. 21–22, 123–125, 201–203; and Claremont Institute, “MGM-140B Block 1A,” http://www
.missilethreat.com/missilesoftheworld/id.74/missile_detail.asp.
60. Most notably, we exclude the Chinese M-9 and M-11 from the Iranian arsenal because neither
has been displayed or tested in public. Although some reports note the rumored deployment of
the systems, others suggest that China at most provided technical assistance on an indigenous Ira-
Furthermore, Iran has used missiles in past operations and often tests its
weapons. In 1994 and 2001, for example, Iran ªred Shahab-1s and -2s at Iraqi
bases used by People’s Mujahideen ªghters to attack Iran.61 More recently,
Iran ªred SRBMs and IRBMs during a series of missile tests and exercises.62
It is unclear whether these tests were successful. Nevertheless, they suggest
that the Iranian military is considering how its missile arsenal factors into its
concept of operations.63
Whereas Iran’s ballistic missile arsenal has become more sophisticated over
time, Saudi Arabian ballistic missile defenses remain relatively limited. Saudi
Arabia relies on the Patriot Advanced Capability-2 (PAC-2) system with ap-
proximately 800 interceptors for ballistic missile defense.64 Standard operating
procedure dictates ªring two interceptors at each incoming missile, giving
Saudi Arabia the ability to target the ªrst 400 missiles before the stockpile is
exhausted.65 Initial reports from the 1991 Gulf War suggested that PAC-2
achieved a 70 percent success rate against Iraqi missiles. Subsequent investiga-
tions, however, indicated that the rate was closer to 10 percent.66 Saudi Arabia
nian project without supplying the missiles. The weapons are also missing from the U.S. Air
Force’s list of Iranian missiles. See Lennox, JSWS, p. 73; Bates Gill, “Chinese Arms Exports to
Iran,” Middle East Review of International Affairs, Vol. 2, No. 2 (May 1998), pp. 60–61; “Iran Missile
Program,” GlobalSecurity.org, May 27, 2005, http://www.globalsecurity.org/wmd/world/iran/
missile-overview.htm; “China’s Missile Exports and Assistance to Iran,” Nuclear Threat Initiative,
September 27, 2005, http://www.nti.org/db/china/miranpos.htm; and National Air and Space In-
telligence Center (NASIC), Ballistic and Cruise Missile Threat (Wright-Patterson AFB, Ohio: NASIC,
U.S. Air Force, April 2009), Report No. NASIC-1031-0985-09, p. 11.
61. Charles P. Vick, “Shahab-2,” Globalsecurity.org, February 1, 2007, http://www.globalsecurity
.org/wmd/world/iran/shahab-2.htm; and “Iran Fires Missiles at Rebel Bases in Iraq,” New York
Times, November 7, 1994.
62. Frederic Wehrey, David E. Thaler, Nora Bensahel, Kim Cragin, Jerrold D. Green, Dalia Dassa
Kaye, Nadia Oweidat, and Jennifer J. Li, Dangerous but Not Omnipotent: Exploring the Reach and Lim-
itations of Iranian Power in the Middle East (Santa Monica, Calif.: RAND, 2009), pp. 64–66; Blanche,
“Iran Stages Display of Missile Firepower”; Robin Hughes, “Iran Launches ‘Great Prophet 2’ Joint
Military Exercise,” Jane’s Defence Weekly, November 8, 2006; Alan Cowell and William J. Broad,
“Iran Reports Missile Test, Drawing Rebuke,” New York Times, July 10, 2008; “Iran Tests Missiles
Amid Nuclear Tension,” CNN.com, September 27, 2009; and “Iran Test-Fires New Version of Fateh
Missile,” CNN.com, August 25, 2010.
63. Blair, “Annual Threat Assessment of the U.S. Intelligence Community for the Senate Select
Committee on Intelligence,” pp. 12–13.
64. Anthony H. Cordesman reports 372 PAC-2 and 432 PAC-2/Guidence Enhanced Munition
(GEM) interceptors. See Cordesman, Saudi Arabia: National Security in a Troubled Region (Santa
Barbara, Calif.: Praeger, 2009), pp. 228, 231.
65. Theodore A. Postol, “Lessons of the Gulf War Experience with Patriot,” International Security,
Vol. 16, No. 3 (Winter 1991/92), p. 145.
66. U.S. General Accounting Ofªce (GAO), “Operation Desert Storm: Data Does Not Exist to Con-
clusively Say How Well Patriot Performed,” GAO/NSIAD-92-340 (Washington, D.C.: GAO, Sep-
tember 1992), pp. 3–4; and Theodore A. Postol, “Lessons of the Gulf War Experience with Patriot,”
International Security, Vol. 16, No. 3 (Winter 1991/92), pp. 119–171. Although the PAC-2/GEM is
designed to improve on the basic PAC-2, we assign it the same 10 percent attrition rate to maxi-
mize Iran’s prospects.
a
Table 2. Iranian Ballistic Missiles, 2009
Range Warhead CEP

Ballistic Missile Type (kilometers) (kilograms) (meters)a Guidance Number
CSS-8/Tondar 69b SRBM 150 250 unknown inertial 200

Fateh A-110c SRBM 210 500 100 inertial-GPS unknown;
initial
operating
capability
in 2004
Shahab-1d SRBM 300 985 450–610 inertial 100–400
Shahab-2e SRBM 500 750–770 700 inertial 100–450
Shahab-3f IRBM 1,300 800 1,850–2,500 inertial ⬎30
Shahab-3Ag IRBM 1,500–1,800 500 1,000 inertial unknown
Shahab-3Bh IRBM 2,000–2,500 500 unknown inertial unknown
Sejil/Ashourai IRBM 2,000 900 unknown unknown unknown
Musudan/BM-25j IRBM 2,500–5,000 1,200 1,600 unknown 18
aCEP refers to the circular error probable of the missile, a measure of accuracy. An SRBM is a short-range ballistic missile and an IRBM is an intermediate-
range ballistic missile. GPS refers to a missile that uses global positioning system in its guidance system.
bDuncan Lennox, ed., Jane’s Strategic Weapons Systems (hereafter JSWS), Vol. 50 (Alexandria, Va.: Jane’s Information Group, 2009), p. 79; and Andrew
Feickert, Iran’s Ballistic Missile Capabilities, CRS Report for Congress (Washington, D.C.: Congressional Research Service, Library of Congress, August 23,
2004), Order Code RS21548, pp. 1–2.
Table 2. (Continued)
cLennox, JSWS, pp. 71–72; and “Fateh-110/NP-110/Mushak,” Globalsecurity.org.

dDomestically produced Scud-B deployed with high-explosive warhead. All data except CEP and stockpile numbers are from Lennox, JSWS, p. 75. CEP and stock-
pile numbers are from Feickert, Iran’s Ballistic Missile Capabilities, pp. 1–3. See also Anthony H. Cordesman and Martin Kleiber, Iran’s Military Forces and
Warfighting Capabilities: The Threat in the Northern Gulf (Washington, D.C.: Center for Strategic and International Studies Press, 2007), pp. 135–139; and
Charles P. Vick, “Iranian Shahab-1/North Korean Scud-B,” GlobalSecurity.org, February 27, 2007. There is disagreement on the number of Shahab-1 missiles in
Iran’s arsenal. Nevertheless, our analysis shows missile requirements far exceed even liberal estimates of Iranian missile stocks.
eDomestically produced Scud-C deployed with HE warhead; Lennox, JSWS, p. 75; Feickert, Iran’s Ballistic Missile Capabilities, pp. 1, 3; and Cordesman and
Kleiber, Iran’s Military Forces and Warªghting Capabilities, p. 135. Note the same dispute described for the Shahab-1 applies to the Shahab-2.
fAlternately named Fajr-3, the Shahab-3 is a domestically produced, near replica of the North Korean No-Dong 1. Lennox, JSWS, pp. 76–78; Feickert, Iran’s Ballistic
Missile Capabilities, pp. 4–5 (for the Shahab-3); Gerald Frost and Irving Lachow, GPS-Aided Guidance for Ballistic Missile Applications (Santa Monica, Calif.:
RAND, 1996), Report No. RP 474-1 (for the No-Dong 1); and National Air and Space Intelligence Center, Ballistic and Cruise Missile Threat (Wright-Patterson AFB,
Ohio: NASIC, U.S. Air Force, April 2009), Report No. NASIC-1031-0985-09, p. 17. The Federation of American Scientists reports better accuracy based on re-
ported developments in the No-Dong program. There is no evidence, however, that such advances have taken place in Iran. Federation of American Scientists,
“Shahab-3/Zelzal-3,” 2008, http://www.fas.org/programs/ssp/man/militarysumfolder/shahab-3.html; “Iran’s Nuclear and Missile Potential: A Joint Threat Assess-
ment by U.S. and Russian Technical Experts” (New York: EastWest Institute, May 2009), pp. 6-12, http://docs.ewi.info/JTA.pdf; and David Holloway and Theodore
Postol, “Comments on ‘Iran’s Ballistic Missile Potential’: A Paper by David Montague, Uzi Rubin, and Dean Wilkening” (New York: EastWest Institute, August
2009), p. 3.
gLennox, JSWS, pp. 76–78.
hIbid.
iIbid., p. 71; Agence France-Press, “Iran Builds New Longer-Range Missile,” November 26, 2007; Robert Wall, “Pentagon Sees Iranian ‘Ashoura’ Missile as
Worrying Development,” November 30, 2007, aviationweek.com; Michael Evans, “Military Experts Say ‘New’ Iranian Missile Is Just an Old One with a Different
Name,” Times Online, November 12, 2008; and Joshua Pollack, “Name That Missile,” online post, November 1, 2009, http://www.armscontrolwonk.com/2519/
name-that-missile.
jThe missile was reportedly delivered from North Korea in 2005. Lennox, JSWS, p. 74.
A Crude Threat 183
has expressed interest in purchasing the more advanced Patriot PAC-3 system,
but the system has not yet been procured.67 Overall, the weakness of Saudi
missile defenses relative to Iranian missile assets suggests the potential attrac-
tiveness of a missile campaign to Iranian planners.
Iran’s missile stockpile consists of a mix of older and newer designs.
Shahab-1s and -2s, domestically produced versions of the Scud-B and Scud-C,
are the most accurate missiles with sufªcient range to reach Abqaiq and the
Petroline pump stations.68 The analysis assumes that Iran would use the
Shahab-1 against these facilities: although the Shahab-1 has a shorter range
than the Shahab-2, its larger warhead and smaller CEP would maximize Iran’s
chance of success.69 Estimates of the size of the Shahab-1 arsenal vary. Our
analysis gives Iran the maximum 400 Shahab-1s in accordance with our worst-
case assumption.70 Although we do not address the Shahab-2 explicitly, the
analysis illustrates the limited effect of 450 additional Shahab-type missiles.
We further assume that Iran would use the more accurate Fateh A-110
against Saudi Arabia’s Persian Gulf oil terminals and stabilization facilities
within the missile’s 210 km range.71 We do not know the size of the Fateh
67. Jim Wolf, “Lockheed Sees Growing Interest in Patriot Missile,” Reuters, August 23, 2007.
68. We believe that there is an important mistake in many reports on Shahab-2 capabilities. The
Federation of American Scientists and Cordesman report that the Shahab-2 has a CEP of 50 meters.
This is a low number given reports in Jane’s, GlobalSecurity.org, and the media suggesting that the
Scud-C (on which the Shahab-2 is based) has a CEP of several hundred meters. We believe that the
lower ªgure results from confusion surrounding Iranian missile programs. In the 1990s, a joint
Syrian-Iranian project produced a missile called the Scud-D modiªed for extended range. During
the Cold War, however, the Soviet Union tested a different missile also labeled Scud-D that em-
ployed optical guidance for a 50-meter CEP; this missile never entered production. We hypothe-
size that researchers saw the Syrian-Iranian project, misattributed the Soviet Scud-D CEP to the
new missile, and subsequently assumed that all Iranian Shahab-2s would enjoy similar accuracy.
See Cordesman and Kleiber, Iran’s Military Forces and Warªghting Capabilities, p. 139; “Shahab-2
(Scud-C),” Federation of American Scientists, February 29, 2008, http://www.fas.org/nuke/
guide/iran/missile/shahab-2.htm; Jane’s Information Group, “Scud Missile,” Jane’s Strategic
Weapons, 2008; and “R-11/SS-1B Scud-A, R-300 9K72 Elbrus/SS-1C Scud B,” Globalsecurity.org, No-
vember 24, 2007, http://www.globalsecurity.org/wmd/world/russia/r-11-specs.htm. We use the
larger CEP because it accords better with reported Scud accuracy and would explain the Iranian
decision to procure newer SRBMs.
69. Estimates of Shahab-1 CEP range from 450 to 610 meters. We use a middle-range estimate of
525 meters. If employing the Shahab-2, Iran would need to ªre more missiles to compensate for its
larger CEP and smaller warhead.
70. Although only Feickert reports 400 missiles, Cordesman and Kleiber report that Iran pur-
chased upwards of 300 Shahab-1s beginning in the mid-1980s and has the ability to manufacture
the Scud-B domestically. It is thus within the realm of possibility that Iran has upwards of 400
Shahab-1s. Feickert, Iran’s Ballistic Missile Capabilities, pp. 1–3; and Cordesman and Kleiber, Iran’s
Military Forces and Warªghting Capabilities, pp. 135–139.
71. Iran tested a longer-range version of the Fateh in September 2010. Because it is unlikely that
Iranian engineers achieved the nearly 50 percent improvement in missile range that would allow it
to reach Abqaiq, this development should not affect our ªndings. See “Iran’s Revolutionary Guard
Gets New Missiles,” Boston Globe, September 21, 2010; and “Iran Test Fires New Version of Fateh
stockpile, although a recent report suggests that the total may be as small as
10.72 Given, however, the extensive testing of Iranian short-range missiles in
recent years, and reports that Iran is producing a successor Fateh, our analysis
assumes a worst case in which Iran has been producing the Fatehs in larger
numbers. Based on Iranian production rates for the Shahab-2, Iran may have
up to 150 Fateh missiles at its disposal. Iran had a maximum of 450 Shahab-2s
at its disposal in 2007, 200 of which were reportedly purchased from North
Korea in the early 1990s.73 The remaining missiles were reportedly produced
domestically after 1997.74 Since the Fateh A-110 reportedly entered production
in 2004 and is—like the Shahab-2—an SRBM, similar procurement rates would
give Iran approximately 150 Fateh A-110s in 2010.
Describing the Attack: Iranian Requirements
This section analyzes the effects of a hypothetical Iranian attack against Saudi
stabilization facilities. Next, it considers the results of an alternative, though mil-
itarily more demanding, attack against Saudi Arabian ports. Finally, it examines
the sensitivity of our ªndings by illustrating how the results change with differ-
ent assumptions regarding Iranian capabilities and Saudi vulnerabilities.
scenario 1: attacking the stabilization facilities

As noted earlier, Abqaiq is the most important stabilization facility in Saudi
Arabia. The heart of the facility consists of eighteen stabilization towers.75 Ten
of the towers are clustered in the northern part of Abqaiq, with the remainder
in the southern half of the facility.76 Cross-referencing Aramco photographs
and descriptions of Abqaiq with Google Earth images of the facility, we believe
Missile,” CNN.com, August 25, 2010. The distances involved require that Iran launch the Fateh
from a small piece of land jutting into the Persian Gulf around 28°00’ N, 51°25’ E.
72. “Strategic Weapons Systems (Iran),” Jane’s Security Sentinel: The Gulf States, May 13, 2009.
73. Cordesman and Kleiber, Iran’s Military Forces and Warªghting Capabilities, pp. 135, 140.
74. Ibid., p. 140.
75. Saudi Aramco, “Abqaiq Plants.” This ªgure contradicts the ten towers reported by Baer and
al-Rodhan. Baer, Sleeping with the Devil, p. xxii; and al-Rodhan, “The Impact of the Abqaiq Attack
on Saudi Energy Security,” p. 3. The confusion might result from the reported presence of ten sta-
bilization towers in the northern part of Abqaiq and additional towers in the southern part of the
facility. See John L. Kennedy, “Giant Abqaiq Complex Handles 60% of Aramco’s Crude Oil,” Oil
and Gas Journal, February 6, 1978. Based on the U.S. experience redeveloping oil infrastructure in
northern Iraq, we believe that each tower operates independently. Like Saudi Arabia, oil from
northern Iraq must be stabilized before export. In 2008, insurgents destroyed two towers at a stabi-
lization plant near Kirkuk. After this attack, however, the undamaged towers continued to oper-
ate. U.S. Army ofªcer, personal communication with authors, November 2009; and James Warden,
“Kirkuk Reªnes Its Oil Industry,” Stars and Stripes, July 15, 2008.
76. Kennedy, “Giant Abqaiq Complex Handles 60% of Aramco’s Crude Oil.”
A Crude Threat 185
that the northern towers fall in a 300 by 20–meter row while the eight southern
towers are in a less linear conªguration.77 The towers themselves have a diam-
eter of less than 6 meters. Less is known about the composition of Saudi
Arabia’s other stabilization facilities at Jubail, Ras al-Juaymah, Ras Tanura, and
Qatif. Using Google Earth, we believe that the only structures within Ras
Tanura matching the Abqaiq towers are within a 20 by 30–meter area.78
We were unable to identify stabilization towers at Qatif, Jubail, and Ras
al-Juaymah; given the similar capacities of the smaller facilities, we use the
measurements of the Ras Tanura towers as a proxy. As discussed above, elimi-
nation of all the stabilization facilities would reduce Saudi exports to 2.6 mbd
of naturally sweet oil.
We assume that a missile striking near the towers could produce sufªcient
overpressure to destroy the towers and their associated machinery. We esti-
mate that it would take 15 pounds per square inch (psi) of peak overpressure
to rupture the stabilization towers.79 Using reference TNT blast curves and
scaling to a 985-kilogram (kg) warhead, the Shahab-1 can produce 15 psi over-
pressure out to a distance of 30 meters.80 Therefore, a missile falling within this
lethal radius of a given tower would cause the tower’s destruction.
To determine the requirements to destroy Abqaiq’s northern towers, we
treat each tower as an aimpoint and send the missiles in salvos of ten (one mis-
sile at each aimpoint).81 Because the towers are close together and the weapons
not perfectly accurate, a missile aimed at one tower could inadvertently de-
77. Saudi Aramco, “Abqaiq Plants”; and Kingdom of Saudi Arabia, Ministry of Petroleum and
Natural Resources, “GAS-ABQ4” in “Photo Gallery: Page 2,” 2006, http://www.mopm.gov.sa/
mopm/detail.do?content⫽photo_gallery2; and Google Earth (accessed 2008–10). Towers are visi-
ble on Google Earth at 25°55’58.00” N, 49°40’59.77” E; 25°55’43.35” N, 49°41’16.95” E.
78. See Google Earth (accessed 2008–10) at 26°41’43.10” N, 50°05’36.32” E.
79. There is no publicly available information on overpressure requirements for stabilization tow-
ers. As a proxy, we used the 15 pounds per square inch (psi) of peak overpressure required to de-
stroy petroleum fractionating towers. See Whitney Raas and Austin Long, “Osirak Redux?
Assessing Israeli Capabilities to Destroy Iranian Nuclear Facilities,” International Security, Vol. 31,
No. 4 (Spring 2007), p. 20 n. 48.
80. The known relationship between overpressure and distance from impact can be scaled to
larger warheads using RL ⫽ D x W1/3, where RL is the lethal radius of the warhead, W is the size of
the warhead in kilograms, and D is the lethal radius for a 1 kg TNT warhead (for 15 psi, approxi-
mately 3 meters). Thus, RL ⫽ 3 x 985(1/3), or 29.8 meters. TNT blast curves taken from U.S. Navy,
“Damage Prediction,” in “Introduction to Naval Weapons Engineering,” Federation of American
Scientists, n.d., http://www.fas.org/man/dod-101/navy/docs/es310/dam_crit/dam_crit.htm.
There is no available information on the actual explosive used in the Shahab warhead. We assume
that it is TNT, recognizing that alternate warheads would produce 15 psi out to a greater distance;
fewer missiles would then be required.
81. Following John Stillion and David T. Orletsky, Airbase Vulnerability to Conventional Cruise-
Missile and Ballistic-Missile Attacks: Technology, Scenarios, and U.S. Air Force Responses (Santa Monica,
Calif.: RAND, 1999), pp. 77–78. The results are not signiªcantly different if aimpoints are placed on
every other tower.
stroy a different one. We thus use a Monte Carlo simulation to produce a ran-
dom landing point for each of the ten missiles according to a circular normal
distribution.82 We then determine whether the random landing point was
within the lethal radius of any tower for each iteration of the simulation. If a
missile lands within the lethal radius of a tower’s center, we consider the
tower destroyed.83
In calculating missile requirements, we assume that Iran wants to be at least
75 percent conªdent that its attack would destroy any given aimpoint.84 The
equations for these calculations appear in note 85.85 The results of the simula-
tion show that Iran would need to launch a minimum of 660 Shahab-1 missiles
to target the ten northern towers even without Patriot attrition. Applied to the
southern towers, the same approach shows that the remaining eight towers re-
quire at least 672 additional missiles. Iran would thus need more than 1,300
missiles to target Abqaiq’s towers with 75 percent conªdence of destroying
any one tower. Including 10 percent Patriot attrition against the ªrst 400 mis-
siles raises the total for Abqaiq to 1,376 Shahab-1s. This does not mean there is
a 75 percent chance all Abqaiq towers will be destroyed by 1,376 missiles.
Rather, given that the probability of destroying each aimpoint is 75 percent,
the likelihood that an attack would destroy all eighteen towers equals 0.75
82. CEP is used to determine the standard deviation of the circular normal distribution around the
aimpoint: CEP/1.1774. We assume that the range and deºection errors are similar, so that
1.1774(σ R + σ D )
CEP ⫽ = 1.1774σ D .
2
See George M. Siouris, Missile Guidance and Control Systems (New York: Springer, 2004), pp. 311–
313.
83. Because a missile could land within the lethal radius of two towers, both towers are consid-
ered “destroyed” by one missile.
84. Seventy-ªve percent is the standard success rate used to calculate munitions requirements by
the U.S. Air Force. Requirements would be different if the Iranians chose a different probability of
success. U.S. Air Force targeting ofªcer, personal communication with authors, April 2008. For
more information on U.S. targeting, see U.S. Air Force, “Intelligence Targeting Guide,” Air Force
Pamphlet No. 14-210, Federation of American Scientists, February 1, 1998, http://www.fas.org/
irp/doddir/usaf/afpam14-210/index.html.
1
85. The overall probability of tower 1 being hit in the ªrst salvo, P hit , is given by
10
1
P hit = 1 − P miss
1
= 1 − ∏( 1 − Pn ),
n=1
1
where P miss is the probability that a missile misses tower 1, and Pn is the probability that tower 1 is
hit by a missile intended for tower n in the 10,000 iterations of the simulation. The average likeli-
n
hood that any given tower is destroyed in a given salvo, Phit, is the average value of P miss across all
ten towers. The number of missiles required to target these towers, N, is thus given by the formula:
log( 1 − Psuccess )
N= ,
log( 1 − P hit )
where Psuccess is the desired probability of success against one aimpoint assigned by the targeteers
(in this case, 75 percent).
A Crude Threat 187
to the eighteenth, or less than 1 percent. If Iran wanted greater conªdence that
its attack would destroy all of the towers, it would need to launch more
missiles.86
Applying the same approach to the other stabilization facilities and substi-
tuting Fateh A-110 speciªcations (500 kg warhead, 100 m CEP) shows that each
facility requires 40 missiles with 10 percent Patriot attrition; destroying all 4 re-
quires at least 160 missiles. Figure 1 summarizes the missile requirements.87
These missile requirements are much lower than for Abqaiq because of the
smaller size of the targets and the Fateh’s greater accuracy.
scenario 2: attacking the export system

Iran might attack Saudi Arabia’s export system as an alternative to targeting
the stabilization facilities. We assume that to stop exports through the Persian
Gulf ports, Iran would have to destroy the machinery and pipes associated
with each tanker berth.88 These berths come in various sizes and shapes.
Table 3 summarizes each target’s size and distance from Iran. Because the Gulf
ports are close to Iranian territory, we assume that they could be targeted with
the Fateh A-110.
Port facilities are signiªcantly more dispersed than the stabilization facilities
and surrounded by water. Given that a water impact would prevent warhead
overpressure from destroying the loading platforms, we assume that Iran’s
missiles would have to directly impact an oil loading point to ensure destruc-
tion.89 Our calculations treat a port target as destroyed if one missile hits
the platform. Missile requirements, calculated using the equations listed in
note 90, for port facilities number in the thousands and are shown in table 4.90
86. For example, Iran would need to launch 3,130 missiles for 50 percent overall success against all
18 towers.
87. The equation is Nneeded ⫽ 0.9(Nlaunched), where Nneeded is the number of missiles required to destroy
a target with 75 percent conªdence and Nlaunched is the number of missiles that must be launched to
overcome Patriot attrition; 0.9 is the proportion of missiles that will reach a target with 10 percent
attrition.
88. Because the pipelines feeding offshore terminals are underwater, Iran would have a limited
ability to destroy them using ballistic missiles. We thus focus on the destruction of the berths.
89. U.S. Air Force targeting ofªcer, personal communication with authors; and U.S. Navy, “Dam-
age Prediction.” This accords with the Iraqi experience at Kharg Island, where only direct hits suc-
cessfully damaged the wharf. See Cordesman and Wagner, “Phase Four: Stalemate and War of
Attrition on Land,” pp. 5–9, 19–21.
90. The probability of a direct hit on each rectangular platform with length L and width W is given
by
2 2
⎛ 1 ⎞ ⎛ L ⎞⎟ ⎛ 1 ⎞ ⎛ W ⎞⎟
P hit = 1 − ( 0.5 )⎜ ⎟ ⎜ × 1 − ( 0.5 )⎜ ⎟ ⎜ .
⎝ π ⎠ ⎝ CEP ⎠ ⎝ π ⎠ ⎝ CEP ⎠
For circular targets with radius r, the probability of a hit is
2
⎛ r ⎞
⎜ ⎟
P hit = 1 − 0.5 ⎝ CEP ⎠ .
Figure 1. Missiles Required to Destroy Stabilization Facilities
Having targeted the Persian Gulf ports, Iran might next attack either the Red
Sea ports or Petroline pump stations to prevent Saudi Arabia from diverting
5 mbd to the Red Sea terminals. In the former case, only Iran’s longest-range
missiles could reach the oil berths at the main Red Sea port of Yanbu.91 We esti-
mate the missile requirements using the warhead size and CEP of the Shahab-
3A.92 Calculations for the destruction of Yanbu’s berths are included in table 4.
91. Yanbu has four crude oil berths and additional berths for reªned products. Using Google
Earth, we believe that each platform measures 35 by 20 meters. We focus solely on crude oil ex-
ports via Yanbu for this portion of our analysis. Saudi Aramco, “Yanbu’ Crude Oil Terminal,”
http://www.saudiaramco.com/irj/portal/anonymous?favlnk⫽%2FSaudiAramcoPublic
%2Fdocs%2FOur⫹Business%2FReªning⫹%26⫹Distribution%2FPorts⫹%26⫹Terminals%2FYanb
u%27&ln⫽en; Saudi Ports Authority, “King Fahad Industrial Port Yanbu,” http://www.ports
.gov.sa/section/full_story.cfm?aid⫽1607&catid⫽20; NGIA, Sailing Directions, pp. 111–112; and
Saudi Aramco, “Yanbu’ Reªnery Marine,” http://www .saudiaramco.com/irj/portal/
anonymous?favlnk⫽%2FSaudiAramcoPublic%2Fdocs%2FOur⫹Business%2FReªning⫹%26
⫹Distribution%2FPorts⫹%26⫹Terminals%2FYanbu%27⫹Reªnery ⫹Marine&ln⫽en.
92. We exclude the smaller Red Sea ports from our analysis, owing to the difªculty of targeting
Yanbu itself.
A Crude Threat 189
Table 3. Targets at Persian Gulf Ports
Distance to Nearest Size of Each

Iranian Territory Target
Port (kilometers)a Targets (meters)b
Ras al-Juaymahc 200 6 mooring buoysd 15 (diameter)

Ras Tanurae 12 berths total
Wharf 200 3 loading platforms 25 x 35
Sea Island 2 200 Sea Island loading platform 40 x 40
Jubailf 200 4 berths 25 x 35g
Zulufh 175 1 mooring buoy 15 (diameter)i
aNational Geographic, Atlas of the World, (New York: Oxford University Press, ), pp. 95–96;
and Google Earth, accessed January 2010.
bSizes estimated using Google Earth (accessed 2008–09).
cSaudi Aramco, “Ju’aymah,” 2010, http://www.saudiaramco.com/irj/portal/anonymous
?favlnk⫽/SaudiAramcoPublic/docs/Our⫹Business/Reªning⫹%26⫹Distribution/Ports⫹
%26⫹Terminals&ln⫽en; and National Geospatial Intelligence Agency (NGIA), Sailing Di-
rections: Red Sea and Persian Gulf, 15th ed. (Bethesda, Md.: U.S. Government Printing
Ofªce, 2009), p. 348.
dEach buoy has one berth.
eEach sea island holds two berths. Less is known of the wharf-based berths, but Google
Earth images from 2004 to 2008 show tankers loading on both sides of the wharf from ap-
proximately three “loading platforms.” Images of Ras Tanura from 2004 to 2008 show
tankers loading on both sides of the wharf from approximately the same location on the
wharf. From this, we infer that each wharf “loading platform” holds two berths. Saudi
Aramco, “Ras Tanura,” 2010, http://www.saudiaramco.com/irj/portal/anonymous?favlnk⫽
%2FSaudiAramcoPublic%2Fdocs%2FOur⫹Business%2FReªning⫹%26⫹Distribution%2FP
orts⫹%261Terminals%2FRas⫹Tanura&ln⫽en; and NGIA, Sailing Directions, pp. 345–347.
fNGIA, Sailing Directions, pp. 349–351, 354; Saudi Ports Authority, “King Fahd Industrial Port
Jubail,” 2003, http://www.ports.gov.sa/section/full_story.cfm?aid⫽1608&catid⫽19; Amy
Myers Jaffe and Jareer Elass, “Saudi Aramco: National Flagship with Global Responsibil-
ities” (Houston, Tex.: James A. Baker III Institute for Public Policy, Rice University, March
2007), p. 86; Anthony H. Cordesman, Saudi Arabia: Guarding the Desert Kingdom (Boul-
der, Colo.: Westview, 1997), p. 85; and M. Webster Ewell, Dagobert Brito, and John Noer,
“An Alternative Pipeline Strategy in the Persian Gulf,” Working Paper (Houston, Tex.:
James A. Baker III Institute for Public Policy, Rice University, March 2007), p. 7, http://
www.rice.edu/energy/publications/docs/
TrendsinMiddleEast_AlternativePipelineStrategy.pdf.
gBecause we are uncertain which facilities at Jubail are used for oil exports, we use the
measurements for the wharf berths at Ras Tanura as a proxy.
hJaffe and Jareer Elass, “Saudi Aramco: National Flagship with Global Responsibilities,”
p. 86; and Cordesman, Saudi Arabia: Guarding the Desert Kingdom, p. 85.
iBecause we cannot locate the buoys at Zuluf, we use buoy measurements for Ras al-
Juaymah as a proxy.
Table 4. Missiles Required to Destroy Port Facilities
Missiles Required
Missiles Required (10% Patriot Attrition
(No Patriot Attrition) against initial 400 missiles)
Port Fateh A-110 Fateh A-110
Ras al-Juaymah 2,958 3,003

Ras Tanura
Wharf 216 240
Sea Island 2 40 45
Sea Island 3 40 45
Sea Island 4 20 23
Jubail, Zuluf, Ras al-Khafji 2,753 2,875
Shahab-3A Shahab-3A
Yanbu 35,904 35,948
Iran could reduce exports via the Red Sea with a successful attack on Saudi
Arabia’s Petroline pump stations. An attack on Pump Station 1 is the most
plausible scenario, because it falls within range of Shahab-1 and -2 missiles.
Although the dimensions of Pump Station 1 are unknown, the locations of
Pump Stations 3, 6, 9, and 10 are in the public record.93 To identify the critical
components of the pump stations, we looked for common features in images
of the known stations. Doing so revealed a common set of ªve buildings cover-
ing an area of approximately 180 by 30 meters. Destroying these targets would
require 660 Shahab-1s with Patriot attrition against the ªrst 400 missiles. In
context, this means that it would take 660 additional missiles to simply reduce
the ºow of oil to the Red Sea, in addition to the more than 6,000 Fateh missiles
required to destroy the Persian Gulf ports.
sensitivity analysis
This section considers how variations in our assessment of Iranian capabilities
and Saudi facilities affect our calculations of Iranian missile requirements. We
analyze the implications of these variations in the following section.
93. Although we cannot ªnd its precise location, Pump Station 1 is reportedly close to Abqaiq.
As the Petroline ºows from Abqaiq to Yanbu, we hypothesize that it is located nearly adjacent
to the Abqaiq facility. Baer, “The Fall of the House of Saud,” p. 54. The locations to Pump
Stations 3, 6, 9, and 10 are known because they have airªelds, the coordinates of which have been
reported to the International Civil Aviation Consortium and posted to GlobalSecurity.org. “Saudi
Airªelds,” GlobalSecurity.org, April 27, 2005, http://www.globalsecurity.org/military/world/
gulf/sa-airªelds.htm. Assuming that Pump Station 1 is adjacent to Abqaiq, the distance from
Pump Station 1 to Pump Station 3 is approximately 240 km. Distances via Google Earth (accessed
2008–10).
A Crude Threat 191
Figure 2. Changes in Missile Requirements for Abqaiq Resulting from CEP Reduction
(no Patriot attrition)
First, the calculations are sensitive to changes in Iranian missile accu-

racy (see ªgure 2). Improving Shahab-1 CEP by 25 percent, which might be
achieved with integrated GPS-inertial guidance in a Scud-type system, results
in a nearly 40 percent decrease in missile requirements.94 Along the same lines,
a hypothetical Fateh with sufªcient range to reach Abqaiq would reduce mis-
sile requirements for its destruction by 90 percent compared to the baseline
Shahab-1 (ªgure 2). This holds despite the Fateh’s smaller warhead. Thus, the
situation facing Saudi Arabia could change dramatically if Iran embarks on a
missile accuracy improvement program.
Relatedly, the calculations are sensitive to changes in the degree to which fa-
cilities are hardened against attack. For example, if stabilization towers cannot
be destroyed by overpressure and instead require a direct hit, then the number
of missiles required for the destruction of Abqaiq with Shahab-1s rises from
1,332 to more than 1 million with no Patriot attrition. Conversely, if the towers
94. Frost and Lachow, GPS-Aided Guidance for Ballistic Missile Applications.
can be destroyed with only 10 psi overpressure, then the lethal radius for the
Shahab-1 warhead grows to 37.8 meters, and missile requirements for Abqaiq
fall from 1,332 to 824.
Effects of an Iranian Attack
The effect of an Iranian attack on the Saudi oil network would depend on the
number and characteristics of available missiles. With current assumptions,
more than 1,300 Shahab-type missiles would be needed to target Abqaiq’s
towers. With the 400 missiles on hand, Iran would be unlikely to do signiªcant
damage. Increasing the desired probability of success raises missile require-
ments: for example, a 50 percent overall probability of destroying Abqaiq’s
towers would require more than 3,300 missiles. Moreover, even if Abqaiq was
destroyed, Saudi Arabia would still be able to produce and stabilize 5.6 mbd of
oil.95 Therefore, even if Iran has many times the number of missiles we esti-
mate, a signiªcant portion of Saudi Arabian oil is secure.
That said, 150 Fateh missiles appear to be sufªcient for Iran to destroy the
smaller stabilization facilities.96 Their destruction would remove an estimated
3.0 mbd stabilization capacity. Because, however, Abqaiq uses only 6.1 of its
13 mbd processing capacity, the destruction of these smaller facilities would
not affect total output. This ªnding holds even if Abqaiq’s capacity is sig-
niªcantly less than estimated. Although we believe that Abqaiq’s actual capac-
ity is much greater, Aramco only reports its capacity as “more than” 7 mbd.97
Using 7 mbd as a lower bound, we calculate that Abqaiq’s spare capacity
could compensate for nearly all 1.25 mbd actually processed by the smaller
stabilization facilities. The remainder could be replaced by additional produc-
tion from Zuluf and Safaniyah. In short, 150 Fateh missiles would be insuf-
ªcient to disrupt Saudi production even with conservative estimates of Saudi
capacity.
Similarly, missile demands are such that an Iranian port attack would be un-
likely to reduce Saudi Arabian exports. If Iran has 150 Fateh missiles, it could
best employ them by targeting Ras Tanura’s Sea Islands, as these facilities han-
dle the port’s largest tankers. Given, however, excess berthing capacity at the
Red Sea and other Persian Gulf ports, a successful attack would have a mini-
95. This amount is based on the 2.6 mbd productive capacity of naturally “sweet” oil and the 3.0
mbd stabilization capacity outside of Abqaiq.
96. Assuming no Patriot interception, Iran requires 144 Fatehs. To eliminate Patriot cover, Iran
could ªre Shahab-type missiles or artillery rockets to “soak up” Saudi interceptors before launch-
ing Fatehs.
97. Saudi Aramco, “Abqaiq Plants.”
A Crude Threat 193
mal effect on exports.98 Overall, a Fateh attack on the Persian Gulf ports could
reduce some of Saudi Arabia’s excess export capacity without degrading ac-
tual exports.
Instead, hundreds of additional missiles would be needed for Iran to
decrease actual exports. Even if the smaller Persian Gulf ports were rendered
inoperable by other means, Iran would need to launch a minimum of 240 mis-
siles at the Ras Tanura wharf in addition to those launched at the Sea Islands.
Alongside a successful strike on the Sea Islands, destruction of the wharf
would eliminate Ras Tanura’s 6 mbd export capacity and leave Saudi Arabian
exports dependent on Ras Juaymah (3 mbd) and the 5 mbd Petroline.99 In this
extreme case, Saudi Arabia could still maintain 8 mbd of exports, only 0.4 mbd
below 2008 levels.
Of course, even if the Iranians were unable to fully destroy the Gulf ports, a
missile attack could lead Saudi Arabia to divert a portion of its exports to the
Red Sea ports. In either case, additional costs would accrue as oil was diverted
to the Red Sea and countries normally serviced from the Gulf encountered
longer transportation times.
What would happen, however, if Iran targeted all 400 Shahab-1s at a portion
of Abqaiq? That is, what would be the maximum damage that the Shahab-1
stockpile could cause? Damage would be minimal, because 400 missiles are
sufªcient to destroy only one tower with a 60 percent chance of success. If
the tower were destroyed, Abqaiq’s total capacity would drop from 13 to
12.3 mbd.100 Because, however, Abqaiq runs below half capacity, its destruc-
tion would have no long-term impact on Saudi Arabia’s ability to stabilize oil.
The facility could still handle its 6.1 mbd throughput with capacity to spare.
Although we do not believe that Iran has sufªcient missiles to take Abqaiq
offline, it is not impossible that Iran will eventually have the capability to do
so. Abqaiq’s destruction would have a signiªcant effect on Saudi exports given
that stabilization is a necessary step for the safe transport of most Saudi oil by
tanker. Taking into account increased production from naturally sweet sources
and assuming the other stabilization facilities could run at full capacity, ap-
98. Destruction of the Sea Islands would eliminate four berths for very large crude carrier-class
(VLCC) tankers and two berths for ultra large crude carrier-class (ULCC) tankers. Excluding
Jeddah, the Red Sea ports have berthing for two ULCC and ªve VLCC tankers; on the Persian
Gulf, Khafji alone has berthing for two VLCC and two Aframax-class tankers other than the six
mooring buoys at Juaymah already used by ULCCs. Thus, Red Sea and Persian Gulf ports could
compensate for losses at Tanura with two ULCC, seven VLCC, and two Aframax berths. See
NGIA, Sailing Directions, pp. 345–355; and Joe Evangelista, ed., “Scaling the Tanker Market,” Sur-
veyor, Winter 2002, p. 6.
99. Because the capacity of the smaller Persian Gulf ports is unknown, we treat them as zero; ac-
tual requirements may be higher.
100. Assuming equal processing, each tower handles 0.7 mbd.
proximately 3.6 mbd could be taken offline by the destruction of Abqaiq. A

maximum of 5.6 mbd, or 61 percent of current production, would then be
available.101
If a successful Iranian attack against Saudi oil facilities occurred, how long
would supply disruptions last? Estimating the time to repair damaged Saudi
facilities depends on the extent of the damage, the ability of manufacturers to
deliver replacement parts, and military-political conditions (e.g., repairs are
likely to take longer if a war is raging). A Congressional Research Service
(CRS) report from the 1970s estimated that signiªcant damage to the Saudi oil
network might take up to a year to repair, given the unique nature of the facili-
ties.102 Repairs to Kuwaiti oil infrastructure after the 1990–91 Persian Gulf War
required two years, with exports resuming within nine months and reaching
83 percent of prewar levels by 1992.103 Likewise, companies reported sig-
niªcant delays in repairing facilities and restarting production nearly two
months after Hurricanes Rita and Katrina.104 Given variation in previous re-
pair times and the uniqueness of Saudi facilities, it is unclear how long it
would take Saudi Arabia to repair the damage caused by a successful missile
attack.
We can, however, estimate the time Saudi Arabia would have to repair the
damage before the loss of Saudi oil affected world oil consumption. As noted,
the loss of Abqaiq would curtail Saudi oil production by 3.6 mbd. In response,
governments around the world could release oil from their strategic petroleum
stockpiles to offset the loss while Saudi Arabia undertook repairs. The U.S.
Strategic Petroleum Reserve alone holds more than 700 million barrels of oil;
the International Energy Agency (IEA) reports total government-controlled oil
reserves of 1.5 billion barrels among IEA members.105 The U.S. stockpile alone
101. Total Saudi production is 9.2 mbd. Aside from Abqaiq, the four stabilization facilities have a
combined capacity of 3 mbd. Naturally sweet production could reach 2.6 mbd. The maximum ca-
pacity taken offline is then equal to 3.6 mbd.
102. Oil Fields as Military Objectives, pp. 1–39. The report notes that Saudi facilities are the largest
of their kind in the world and require specialized production of custom equipment. Similarly, Baer
postulates that destruction of Abqaiq would take “months” to repair. Baer, “The Fall of the House
of Saud,” p. 54.
103. John Cranªeld, “Fires Are Out, but Work Continues,” Petroleum Economist, December 30,
1991, p. 17; and Bob Tippee, “Kuwait Pressing toward Preinvasion Oil Production Capacity,” Oil
and Gas Journal, March 15, 1993, p. 41; Kuwait produced 1.057 mbd in 1992 against 1.278 mbd in
1989. OPEC, Annual Statistical Bulletin, 2008, p. 21.
104. “Hurricanes Katrina and Rita,” hearings before the Senate Committee on Energy and Natural
Resources, 109th Cong., 1st sess., S. Hrg. 109-279 (Washington, D.C.: U.S. Government Printing
Ofªce, October 27, 2005), pp. 8–18.
105. International Energy Agency, “IEA Response System for Global Supply Emergencies” (Paris:
International Energy Agency, 2007), p. 7; and United States Department of Energy, “Strategic Pe-
troleum Reserve Inventory,” November 30, 2010, http://www.spr.doe.gov/dir/dir.html. This
ªgure does not take into account private and non-IEA government stockpiles.
A Crude Threat 195
would be sufªcient to offset the losses from the destruction of Abqaiq for more
than six months, given 2008 consumption rates. If all 1.4 billion barrels in gov-
ernment reserves were employed, the repair window would be nearly ªfteen
months.106 A six-to-ªfteen-month window is in the midrange of past repair ex-
periences, suggesting that world oil consumption would not be impaired even
after a successful Iranian attack.
Potential Responses to an Iranian Attack
This section brieºy discusses a range of potential Saudi Arabian responses to

an Iranian attack on its oil facilities. These options, however, are either costly
or of limited likely effectiveness.
First, Saudi Arabia might consider procuring better missile defenses. Unfor-
tunately, this is an expensive proposition. Even if Saudi Arabia procured the
improved Patriot Advanced Capability 3 (PAC-3) system to replace the PAC-2,
it is unclear whether the cost-beneªt analysis would be in Saudi Arabia’s fa-
vor. Standard operating procedure for the Patriot system speciªes ªring two
interceptors at each incoming missile. As a result, each increment of growth in
the Iranian arsenal requires a disproportionately large Saudi investment.
Second, Saudi Arabia might attempt to attack Iranian missile launchers to
limit missile strikes. Even an unsuccessful anti-launcher campaign could re-
duce attacks as launchers are moved to avoid detection. The small area from
which Iran can launch its most accurate missiles against Saudi Arabia would
allow the Saudis to concentrate their efforts.107 Iran, however, could simply re-
spond by concentrating air defenses and other assets to protect launchers in
the area. Furthermore, missile hunting has an inauspicious history: despite
thousands of sorties, the U.S.-led coalition may not have destroyed a single
Iraqi Scud launcher throughout Operation Desert Storm. These factors suggest
a Saudi antimissile campaign would face severe limitations.108
Finally, and to the extent Iran employs GPS guidance in its missiles, Saudi
Arabia might consider jamming GPS signals. GPS jamming, however, is of
questionable utility. There are two types of GPS jamming.109 With “smart” jam-
ming, Saudi Arabia would erect emitters to create false GPS signals and divert
106. This period is calculated by dividing the size of global reserves by lost Saudi production.
These windows may be artiªcially narrow because Saudi production may gradually return rather
than suddenly restart.
107. See note 71.
108. Mark E. Kipphut, “Crossbow and Gulf War Counter-Scud Efforts: Lessons from History,”
Counterproliferation Papers, Future Warfare Series, No. 15 (Maxwell Air Force Base, Ala.: USAF
Counterproliferation Center, February 2003).
109. Scott Pace, Gerald Frost, Irving Lachow, David Frelinger, Donna Fossum, Donald K. Wassem,
Iranian missiles away from their targets.110 This form of jamming, however, is
most effective before GPS receivers lock on to the actual GPS signal. Saudi
Arabia would therefore need to erect emitters on Iranian territory. This is not
an impossible task, but Iran could move its missiles after it notices them re-
peatedly missing their targets.
The second option is “noise” jamming. With noise jamming, Saudi Arabia
would block all GPS signals over a predetermined area. When a missile enters
the jammed area, the loss of GPS forces it to rely solely on inertial navigation.
Inertial guidance introduces navigation errors resulting from guidance drift,
so the more time a missile spends in the jammed area, the less accurate it be-
comes. Because even a moderately powered noise jammer can block GPS sig-
nals out to more than 100 km, a Scud-type missile traveling at 1.5 km per
second would then spend more than 1 minute in the jammed area, allowing
time for signiªcant errors to accrue.111 Noise jamming, however, cannot dis-
criminate friend from foe. Therefore, it would also disrupt Saudi GPS systems
and hinder Saudi civilian and military operations. Noise jamming thus has
signiªcant drawbacks.
In sum, the limits of missile defense, missile hunting, and GPS jamming sug-
gest that Saudi Arabia could not easily prevent an Iranian attack. Further in-
vestment in these areas would be costly and would not provide a large return.
Instead, given the problems associated with disrupting an Iranian attack,
Saudi Arabia might consider efforts to mitigate potential damage. Although
Iran is unlikely to cause signiªcant damage with its current missile assets,
damage mitigation would become important if Iran increases missile accuracy.
Our analysis suggests, therefore, that Saudi Arabia should focus on improving
and expanding its stabilization and port facilities.
If Saudi Arabia believes that a missile campaign is likely, then it should
seek to reduce vulnerabilities in stabilization infrastructure. Stabilization is
an essential step in safely exporting Saudi oil, yet the vast majority of
Saudi Arabia’s current capacity is concentrated in a small area. Saudi Arabia
could reduce its vulnerability by building additional stabilization facilities
around the kingdom and hardening existing infrastructure.112
and Monica Pinto, The Global Positioning System: Assessing National Policies (Santa Monica, Calif.:
RAND, 1995), p. 48.
110. Ibid., pp. 44–45; and Gerald Frost, Operational Issues for GPS-Aided Precision Guided Weapons
(Santa Monica, Calif.: RAND, 1994), p. 27.
111. Pace et al., The Global Positioning System, pp. 52–53. For Scud velocity, see Charles P. Vick, “Ira-
nian Shahab-1/North Korean Scud-B,” GlobalSecurity.org, February 27, 2007, http://www
.globalsecurity.org/wmd/world/iran/shahab-1-specs.htm; U.S. Air Force ofªcer, personal com-
munication with authors, January 2010; and Frost, Operational Issues for GPS-Aided Precision Guided
Weapons, pp. 27–28.
112. Warrick, “U.S. Steps Up Arms Sales to Persian Gulf Allies.”
A Crude Threat 197
Alternatively, if Saudi Arabia believes that a port attack is likely, it should

stockpile ºoating mooring buoys to replace destroyed berths. This effort
would replicate Saudi steps toward pipeline security, where replacement sec-
tions of pipeline are stockpiled beforehand. It would also mirror Iraq’s efforts
to compensate for damage sustained during the Iran-Iraq War.113
Implications for Other Scenarios
This section brieºy reviews the implications of our ªndings for other Iranian
retaliatory options—speciªcally, a conventional air campaign, naval assault,
and attacks by special forces and proxy groups—and what they mean for fu-
ture analyses. Although more work is needed to develop these scenarios,
this study suggests that Saudi Arabia and the international community should
not be overly concerned with other Iranian options for disrupting Saudi oil
production.
conventional air campaign

To disrupt Saudi Arabian oil production using a conventional air campaign,
the Iranian air force (IRAF) would likely attack the stabilization and port facili-
ties described above. Although IRAF can nominally ªeld 118 ªghter and 168
attack aircraft, large numbers of IRAF aircraft are obsolescent. Many are no
longer operational, with IISS reporting 60 percent serviceability for Western
aircraft and 80 percent for Soviet-Chinese types. In attacking Saudi Arabia, the
IRAF could thus deploy approximately 66 ªghters and 115 attack aircraft.114
These would have to overcome as many as 126 modern Saudi ªghters and
long-range surface-to-air missiles.115 Assuming this was successful, the IRAF’s
115 attack aircraft would then have to overcome remaining Saudi Arabian
ªghters and air defenses to attack several dispersed targets.116
The preceding scenario seems an unlikely assignment for the IRAF because
the numbers and qualitative edge are in Saudi Arabia’s favor. Although we
cannot discount the possibility that Iran would launch such an attack in extre-
113. “Persian Gulf Oil Exporters Brace for Next Step in Iran/Iraq War,” Oil and Gas Journal, De-
cember 12, 1983, p. 41.
114. Authors’ calculations, based on International Institute for Strategic Studies (IISS), The Military
Balance, 2008 (London: IISS, 2008), pp. 243–244; and Cordesman and Kleiber, Iran’s Military Forces
and Warªghting Capabilities, p. 95.
115. “Royal Saudi Air Force,” GlobalSecurity.org, April 27, 2005, http://www.globalsecurity.org/
military/world/gulf/rsaf.htm; IISS, Military Balance, 2008, p. 261; International Institute for Strate-
gic Studies, The Military Balance, 2009 (London: IISS, 2009), p. 265; and James C. O’Halloran, ed.,
Jane’s Land-Based Air Defence, 2008–2009 (Alexandria, Va.: Jane’s Information Group, 2008), pp.
403–404, 410–411.
116. IISS, The Military Balance, 2008, p. 262; O’Hallorran, Jane’s Land-Based Air Defence, 2008–2009,
pp. 140, 417–418, 481–482; and Cordesman and Obaid, National Security in Saudi Arabia, p. 247.
mis, the relative balance of power suggests the questionable utility of a con-
ventional air campaign. Still, if Saudi Arabia is worried about an Iranian air
campaign, it could procure additional air defenses to deploy along avenues of
potential attack and improve the readiness of the Royal Saudi Air Force.
naval assault
Attacks by the Iranian navy would presumably target oil terminals along the
Gulf coast. Iran conducted similar operations during the Iran-Iraq War when
naval forces attacked the Iraqi Persian Gulf oil terminals.117 The Iranian navy
has several frigates and corvettes, and more than 100 smaller craft that could
be used in an attack.118
To substantially reduce Saudi exports, however, Iran would need to destroy
upwards of two dozen berths and buoys. Destroying such a large number of
targets would be complicated by the small size and geographic dispersion of
these targets. Likewise, unless an attack was well coordinated, initial assaults
would alert the Saudi military and enable it to improve security at other ports.
Even if these attacks were successful, Saudi Arabia could continue to export at
least 5 mbd via the Red Sea ports. Meanwhile, unless a blockade was main-
tained around the Persian Gulf ports, Saudi Arabia could replace destroyed fa-
cilities with mooring buoys. In short, the large number of targets, excess
network capacity, and Saudi repair options would limit the consequences of a
naval attack.
Still, if Saudi Arabia is worried about an Iranian naval assault, the preceding
analysis suggests that it could take two relatively easy steps to mitigate the
consequences. First, it could stockpile oil on the Red Sea coast to take advan-
tage of slack capacity at the Red Sea terminals. Second, it could stockpile
mooring buoys and other parts to replace damaged facilities.
attacks by special forces and proxy groups

Iran might also consider using special forces or proxy groups to attack
the Saudi oil network.119 It is difªcult to estimate the likelihood of success for
these attacks. While Iran has reportedly built up its special forces and ex-
pressed interest in supporting local proxies, Saudi Arabia has recently taken
steps to improve forces devoted to oil security; we cannot say which side
would succeed in its mission.120 Our analysis of Saudi infrastructure, however,
117. Hiro, The Longest War, p. 47.

118. International Institute for Strategic Studies, The Military Balance, 2010 (London: IISS, 2010),
pp. 251–253.
119. Cordesman, “The United States, Israel, the Arab States, and a Nuclear Iran,” pp. 100–120.
120. EIA, “Country Analysis Brief: Saudi Arabia,” November 2009, p. 6; and Warrick, “U.S. Steps
Up Arms Sales to Persian Gulf Allies.”
A Crude Threat 199
suggests that attackers would likely face substantial difªculties because targets
are widely dispersed and physically robust. For instance, to successfully inca-
pacitate Abqaiq, an attacker would need to penetrate Saudi defenses and then
destroy at least ten of the stabilization towers to lower Abqaiq’s capacity
below 2009 processing levels. Given the size and physical durability of the
towers, this seems a difªcult proposition for lightly armed assault forces.
Moreover, Saudi security forces would be trying to defeat an attack through-
out, adding another layer of complexity to an operation. Still, if Saudi Arabia
worries about an asymmetric attack, it should continue improving its oil secu-
rity forces and consider adding more redundancies to the oil network.
Conclusion
Overall, we judge the Saudi oil network to be secure against an Iranian missile
attack given existing Iranian capabilities. Current Iranian missile holdings are
insufªcient in number and quality to destroy the stabilization facilities that
would cause the greatest reduction in Saudi oil production. Although an
Iranian missile attack could reduce some export capacity at the Persian Gulf oil
ports, the lost capacity could be replaced by excess capacity at the Red Sea and
smaller Gulf terminals. Dispersion and redundancies in the Saudi oil network
make the rest of Saudi production difªcult to incapacitate. In short, this analy-
sis shows that a missile campaign is not a proximate military threat.
There is an important caveat to this assessment: although an Iranian missile
campaign is unlikely to physically disrupt the ºow of Saudi oil, any Iranian at-
tack would likely have a signiªcant impact on the world price of oil. A failed
2006 terrorist attack against Abqaiq, for instance, caused no damage to the fa-
cility, yet caused a temporary $2 increase in the per barrel price of oil.121 We
also cannot say whether shipping companies would continue tanker opera-
tions in a war zone; at a minimum, insurance rates for tankers transiting the re-
gion would increase. Thus, even a failed missile campaign could raise oil
prices for consumers. Although we cannot predict the speciªc effect of an at-
tack on market prices, it would likely be out of proportion to the physical dam-
age inºicted. Still, because our analysis indicates the actual damage of an
Iranian campaign would be minimal, these price shocks would be based on in-
complete information, short-term uncertainty, and speculation. As the limited
efªcacy of Iranian attacks became clear, and if the international community
worked to calm markets by, for instance, opening strategic petroleum reserves,
prices should eventually fall.
121. Al-Rodhan, “The Impact of the Abqaiq Attack on Saudi Energy Security.”
More generally, our analysis carries implications for understanding devel-

opments in the Iranian ballistic missile arsenal, energy security around the
Persian Gulf, and regional security dynamics. First, our analysis adds nuance
to the debate over Iranian missile capabilities. Whereas analysts in the United
States often highlight the increasing range of Iranian missiles, the more worri-
some development from the standpoint of regional security would be Iran’s
acquisition of increasingly accurate missiles.122 Even moderate gains in
accuracy—for example, improving Shahab-1 CEP by 25 percent—result in siz-
able reductions in the number of missiles required to destroy a facility. There-
fore, sustained improvements in the accuracy of Iranian ballistic missiles
would enable Iran to do signiªcantly greater damage with an arsenal of a
given size. Evidence that Iran has made the technological leap to designing
missiles fully able to exploit the gains in accuracy from GPS-based guidance
would be the most worrisome: at that point, Iran would be able to disrupt oil
production even with a small arsenal.123 In the future, technology transfer
from more advanced nations, combined with greater sophistication within the
Iranian program itself, could lead to more effective GPS employment. If Iran is
found to be on an accuracy improvement curve, then it would be prudent for
states around the Persian Gulf to harden facilities or add further redundancies
to their oil networks; missile defenses, though costly, may also become an at-
tractive option.
Second, this analysis can inform the international community’s efforts to
confront Iran over its nuclear program. Since the extent of the Iranian program
ªrst came to light, policymakers and analysts have considered using force to
restrain Iran’s nuclear ambitions. This debate weighs the prospective gains in
slowing the Iranian program with the potential costs paid given Iranian retali-
ation. Our analysis demonstrates that Iranian retaliation is unlikely to cause
real losses to oil supplies. There are still real reasons why the United States
may not want to strike Iran—for instance, Iran could retaliate by arming prox-
ies to ªght U.S. forces in Afghanistan. Nevertheless, the missile threat to oil
does not constitute a credible Iranian deterrent and, at least in a military sense,
122. See Hildreth, Iran’s Ballistic Missile Programs; NASIC, Ballistic and Cruise Missile Threat, pp. 3,
14, 18; Blair, “Annual Threat Assessment of the U.S. Intelligence Community for the Senate Select
Committee on Intelligence,” p. 13; Associated Press, “Pentagon Studying Intel on Iranian Missile
Test Launch as Congress Mulls Action,” FoxNews.com, July 9, 2008; and Robert D. Walpole, “Ira-
nian Ballistic Missile, WMD Threat to the U.S.,” statement for the record to the International Secu-
rity, Proliferation, and Federal Services Subcommittee of the Senate Government Affairs
Committee, 109th Cong., 2d sess., September 21, 2000.
123. In addition to employing GPS guidance, such a missile must be able to mechanically respond
to GPS guidance updates. Gerald Frost and Irving Lachow, GPS-Aided Guidance for Ballistic Missile
Applications (Santa Monica, Calif.: RAND, 1996), Report No. RP 474-1.
A Crude Threat 201
concerns about Iranian retaliation against oil production should not factor
signiªcantly in the debate.
Third, this study holds general implications for U.S. policy in the Persian
Gulf. Since President Jimmy Carter’s administration, the United States has
committed itself to maintaining the free ºow of oil from the Persian Gulf. To
achieve this aim, the United States has made a substantial military effort to de-
fend the region, including ªghting one major war in 1991 and devoting billions
to military spending.124 Our ªndings challenge part of the rationale for this ef-
fort. In showing that the Saudi oil network would be resilient in the face of a
concerted attack by one of the most capable actors in the region, our research
indicates that threats to regional oil production are overblown. By implica-
tion, the United States may be able to reduce its military commitment to the
region.125
Finally, this article carries implications for understanding the vulnerability
of oil infrastructure and oil networks. Simply put, oil is not an easy military
target. Individual production facilities are large and physically robust, requir-
ing a signiªcant military effort to disable. Facilities are also geographically dis-
persed, necessitating systematic targeting. Given the presence of redundant
facilities, some oil networks may have few, if any, targets that can incapacitate
an entire system. Even if there are critical nodes, states can add facilities
to limit the vulnerability of an oil network to disruption. If an attack is feared,
meanwhile, states can stockpile replacement equipment to mitigate pros-
pective damage. In sum, oil is a lucrative target, but it is not universally
vulnerable.
124. Mark A. Deluchi and James J. Murphy estimate that the United States spends $30 to $60 bil-
lion per year to defend its interests in the Persian Gulf. Delucchi and Murphy, “U.S. Military Ex-
penditures to Protect the Use of Persian Gulf Oil for Motor Vehicles” (Davis: Institute of
Transportation Studies, University of California, Davis, March 2008), Report UCD-ITS-RR-93-3,
p. 15.
125. For more on this point, see Eugene Gholz and Daryl Press, “Footprints in the Sand,” American
Interest, Vol. 5, No. 4 (March–April 2010), pp. 59–67.

A Crude Threat:The Limits of An Iranian Missile Campaign Against Saudi Arabian Oil

Uploaded by

Copyright:

Available Formats

A Crude Threat:The Limits of An Iranian Missile Campaign Against Saudi Arabian Oil

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Crude Threat:The Limits of An Iranian Missile Campaign Against Saudi Arabian Oil

Uploaded by

Copyright:

Available Formats

A Crude Threat :The Limits of an Iranian Missile Campaign

against Saudi Arabian Oil

Citation Itzkowitz Shifrinson, Joshua R. and Miranda Priebe. “A Crude

Version Final published version

A Crude Threat Joshua R. Itzkowitz

International Security, Vol. 36, No. 1 (Summer 2011), pp. 167–201

oil installations—speciªcally, in the scenario below, Saudi Arabian facilities—

12. Talmadge, “Closing Time,” especially p. 115.

taliation. The article concludes by discussing the implications of our analysis

Setting and Assumptions

We do not consider an unprovoked Iranian attack on Saudi Arabian oil infra-

Saudi Arabia’s Oil Infrastructure and Its Vulnerabilities

From a stabilization plant, crude oil is pumped either directly to a port

Table 1. Overview of Saudi Arabian Production and Facilities

Productive capacity ⬎ 11.8

Iranian and Saudi Arabian Forces

Range Warhead CEP

CSS-8/Tondar 69b SRBM 150 250 unknown inertial 200

cLennox, JSWS, pp. 71–72; and “Fateh-110/NP-110/Mushak,” Globalsecurity.org.

Describing the Attack: Iranian Requirements

scenario 1: attacking the stabilization facilities

scenario 2: attacking the export system

Figure 1. Missiles Required to Destroy Stabilization Facilities

Table 3. Targets at Persian Gulf Ports

Distance to Nearest Size of Each

Ras al-Juaymahc 200 6 mooring buoysd 15 (diameter)

Table 4. Missiles Required to Destroy Port Facilities

Ras al-Juaymah 2,958 3,003

First, the calculations are sensitive to changes in Iranian missile accu-

Effects of an Iranian Attack

proximately 3.6 mbd could be taken offline by the destruction of Abqaiq. A

Potential Responses to an Iranian Attack

This section brieºy discusses a range of potential Saudi Arabian responses to

Alternatively, if Saudi Arabia believes that a port attack is likely, it should

Implications for Other Scenarios

conventional air campaign

attacks by special forces and proxy groups

117. Hiro, The Longest War, p. 47.

More generally, our analysis carries implications for understanding devel-

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.