In a surprise move this year, the Pentagon has reduced spending for two aerial drones. A version of the RQ-4 Global Hawk will be relegated to storage to be superseded by more capable versions, and future production of the MQ-9 Reaper is to be reduced from 48 per year to 24.
The decisions were surprising. Drones are widely touted as the future of warfare. How can it be that the 40-year old, manned U-2 reconnaissance aircraft can do the mission better than even an early-generation drone?
The Reaper decision was not attached to any admission of disappointment; it was just a matter of budget constraints and skilled manpower shortages, DOD said. The minor setback with Global Hawk notwithstanding, the aura of a leap-ahead in war-fighting technology is left intact, or so it is to be believed.
Much has been written about unmanned aerial drones. Some of it has questioned the morality of how they are being used, and in a few cases, some aspects of technical performance is questioned. Much more of the writing and the vast majority of expert opinion is that drones are cheaper to buy and operate than manned aircraft, can do things aircraft cannot do, and when they perform aircraft-type missions, they often do them at least as well, if not betterall without endangering an American pilot. Some even proclaim the F-35 Joint Strike Fighter to be the Air Forces last manned tactical aircraft, and the Air Force is seriously contemplating an optionally manned long range, nuclear bomber.
For example, Peter Singer of the Brookings Institution argues that an amazing revolution is taking place on the battlefield; drones will increasingly be available as autonomous force multipliers. Singers colleague at Brookings, Michael OHanlon, argues that The era of manned airplanes should be seen as over, and they both argue that those who persist in supporting manned systems are standing in the way of progress in defense technology, if not history. Even some promoters of manned systems tend to agree for the long term; one advocate of all things Air Force, John A. Tirpak, executive editor of Air Force Magazine, favorably compares the Reaper drone to the F-16 as a bomber, and retired Air Force General David Deptula, often quoted by the media on drone topics, has argued that the next step is to reduce, if not eliminate, the role of humans in operating drones, even from the ground.
The defense trade press assiduously reports advances in drone technology; Congress has called for one-third of the aircraft in the operational deep strike force aircraft fleet to be unmanned, and there is a bipartisan Unmanned Systems Caucus in Congress, co-chaired by the ubiquitous chairman of the House Armed Services Committee, Cong. Buck McKeon (R-CA).
All of it gives the aura of inevitability to the future of drones and an impression of cutting edge insight to those who opine on what should seem obvious to thinking, informed analysts and commentators.
The rhetoric harks back to the revolution in military affairs proclaimed shortly before the first Gulf War (Operation Desert Storm) in 1991, the one bomb one target precision affirmed after that war, and the shock and awe that experts predicted would implode Saddam Husseins regime in a matter of hours at the start of the 2003 war.
This author has some experience with such prognostications. From 1992 to 1996, I worked with a team in the Government Accountability Office (GAO) to assess the effectiveness of the air war in Operation Desert Storm. It was not one bomb one target; for bridges, for example, it was an average of eleven laser guided bombs to make any bridge un-useable; for other targets it was more. Other data show that precision attacks on Saddam Husseins air defenses on the first night of Operation Desert Storm and on Saddam Hussein himself in Operation Iraqi Freedom in 2003 definitively failed to achieve the intended objectives and frequently missed their aim points.
Todays augury on drones makes the same facile predictions and uses the same rhetoric. It has meant more money. In the twelve years before the 9/11 attacks (1988-2000), DOD spent just $3.9 billion for drones. After 9/11, spending increased dramatically, from $667 million in 2001 to $5.1 billion in 2011  in known  drone costs, totaling $30 billion. CBO estimates that spending will increase again in the next decade, to $37 billion. But even that estimate will probably turn out to be low, the new downturns announced for Global Hawk and Reaper notwithstanding.
Amidst the enthusiasm for drones and the increasing money, there has been little public diagnostic analysis of what a specific drone can and cannot do, how well or poorly, the cost, and how it all compares to relevant comparison manned aircraft. A recent and informative report from the Congressional Research Service (CRS) suggests at least three areas of comparison: unmanned aerial systems eliminate the risk to a pilots life, and their aeronautical capabilities, such as endurance, are not bound by human limitations.... [They] may also be cheaper to procure and operate than manned aircraft.
Selection of Reaper
The Air Forces MQ-9 Reaper is selected for analysis. In combat since October 2007, it has been employed long enough to have an operational record for a reality-based analysis, and as a successor to the earlier MQ-1B Predator (deployed in 2002 as the first U.S. armed aerial drone in the modern age ), it is not so new to be dismissed, or defended, as too embryonic for meaningful analysis.
Various websites provide a physical description of Reaper. These include sites that convey some useful basic information but also tend to describe performance as uncompromised by limitations and planned upgrades as fully in hand; sometimes important data is completely missing. Such sites include those of the producer, General Atomics, GlobalSecurity.org, and the Air Force. The Wikipedia entry for Reaper provides more details and links, but it also gives the impression of being written by advocates. Others, such as the Government Accountability Office (GAO) and the Defense Departments Director of Operational Test and Evaluation (DOT&E), are more objective, but they are also cryptic. More complete descriptions and analysis are provided by the Congressional Research Service  and the Congressional Budget Office (CBO); however, even the more informative of these (CRSs) only touches on issues that when scratched can be quite revealing.
The first point that almost all descriptions make about Reaper is that it is not a stand-alone, pilot-less aircraft. It is a system with elements that go beyond what piloted aircraft require. A Reaper unit does not consist of one aircraft; it consists of four (why will be explained in subsequent parts), and it requires a manned ground control station (GCS) that is typically remotely located, often in the US. It also requires a dedicated Primary Satellite Link (PSL) and a local control unit for landings and take-offs where the air vehicles are based. In comparing drones to manned aircraft, CRS points out, for example, it is necessary to include these components that are essential for flight.
Unclassified DOD records show Reaper spending to have started in 2002. Congress quickly became a major advocate and increased the purchase by 58 air vehicles. Production had been planned to terminate at the end of 2016 at 399 individual air vehicles, and total Reaper spending had been projected at $12.497 billion in nominal (then year) dollars. The 2013 budget request included the reduction in Reapers production rate from 48 to 24 per year for the years 2013-2017, which would imply a reduction in the total buy by either 96 or 72 air vehicles. It is also not clear how much this change would reduce the cost of the total buy; unit costs will go up with the rate change, and one budget document addressed buying more ground control stations with the money that is saved from the reduced production.
In any case, the total amount for Reaper costs is not borne by the Reaper program alone: Reaper uses the same ground equipment as Predator B; there may be GCS and other ground equipment paid for by the Predator B program but employed by Reapers, and more than one DOD official informally speculated that other Reaper costs may be buried in other programs, such as Gorgon Stare.
Distinguishing Between Reapers
Important to appreciate about Reaper is that it is not a Predator, nor a minor modification of it as the producers moniker, Predator B, might lead one to believe. The Air Force makes the distinction between them clear in its factsheets. While they dissemble on issues like cost and performance, they can at least be taken at face value on some basics:
|Characteristic||MQ-1B Predator||MQ-9 Reaper|
|Wingspan||55 ft.||66 ft. or 86 ft.|
|Empty Weight||1,130 lbs.||4,900 lbs. for 66 ft. wingspan variant|
|Maximum Take Off Weight||2,250 lbs. with up to 665 lbs. of fuel and 450 lbs. of weapons||10,500 lbs. with up to 4,000 lbs. of fuel and up to 3,000 lbs. of external wing stores (fuel and/or weapons) and a 750 lb. internal sensor payload. Any payload combination may not exceed 5,600 lbs.|
|Sensor Package||Multi-spectral Targeting System (MTS-A) with infrared sensor, daylight TV camera with image-intensifier & laser illuminator/designator||MTS-B with basically same sensors, plus synthetic aperture radar (SAR) to enable GBU-38 JDAM targeting. (See SAR/JDAM discussion below.)|
|Armament||2 laser-guided AGM-114 Hellfire missiles||Hellfire and/or 500 lb. GBU-12 laser guided bombs up to a total of 3,000 lbs.|
|Speed (kt)||70 (cruise/loiter); 118 (max.)||120 (cruise/loiter); 240 (max.)|
|Range||Up to 770 miles||Up to 1,150 miles|
|Ceiling||Up to 25,000 ft. depending on payload||Up to 50,000 ft. depending on variant and payload|
|Endurance||Up to 40 hours; 14-16 hours with wing stores||Up to 24 to 30 hours with no external stores; 14 hours with typical munitions; up to 42 hours with 2 wing fuel tanks and 1,000 lbs. of weapons|
|Initially Deployed||March 2005||October 2007|
Note that, despite some similarity in appearance, Reaper is an entirely different air vehicle from the MQ-1B Predator (or the original, unarmed RQ-1A Predator). At 4,900 pounds, Reapers empty weight is four times that of Predator, and its wingspan is 20 to 56 percent wider. Reaper has loiter and maximum speeds that are roughly twice Predators, but both are slow compared, for example, to modern combat aircraft. (The slow speed is a real advantage for finding and identifying targets.) Perhaps the most noticeable difference in the table above is the much larger payload of weapons that Reaper can carry, up to 3,000 pounds rather than Predator Bs very modest 450 pounds.
There are also important similarities. Both can loiter above the battlefield searching for targets for long periods (up to 40 hours), much longer than manned aircrafteven when the latter can be refueled in the air. Similarly, loiter endurance is significantly reduced depending on the weight and drag of munitions, down to 14-16 hours, perhaps less. The two may differ little in their maximum ceiling with no or light payloads, but there is a high altitude, longer wingspan version of Reaper for as high as 50,000 feet claimed.
Next: Cost and Performance
 Several titles have been applied to drones: unmanned aerial systems (UASs), unmanned aerial vehicles (UAVs), and remotely piloted vehicles (RPAs) are the most common.
 For example, see David Cortrights License to Kill at http://www.cato-unbound.org/2012/01/09/david-cortright/license-to-kill/, the work of David S. Cloud at the LA Times, such as at http://articles.latimes.com/2011/dec/29/world/la-fg-drones-civilians-20111230, work at the U.K.s The Guardian, such as at http://www.guardian.co.uk/commentisfree/cifamerica/2011/nov/07/cia-unaccountable-drone-war or at http://www.guardian.co.uk/world/2011/jul/17/us-drone-strikes-pakistan-waziristan. There is also an excellent summary of the aftereffects of drones and occupation in Iraq at Salon.com. http://www.salon.com/2012/01/30/lessons_from_iraqi_outrage_over_us_drones/singleton/. See also Nick Turses work at TomDispath.com at http://www.tomdispatch.com/post/175489/tomgram%3A_nick_turse%2C_drone_disasters_/ and http://www.tomdispatch.com/post/175482/tomgram%3A_nick_turse%2C_the_life_and_death_of_american_drones/
 See the website for P.W. Singers Wired for War at http://wiredforwar.pwsinger.com/.
 See With more details coming, analysts split on new DOD strategy, Chris Carroll & Leo Shane III, Stars and Stripes, January 25, 2012 at http://www.stripes.com/news/with-more-details-coming-analysts-split-on-new-dod-strategy-1.166755.
 See Rest in Peace, Manned Aircraft, Air Force Times, April 16, 2011, posted by David Larter at http://militarytimes.com/blogs/flightlines/2011/08/16/rest-in-peace-manned-aircraft/.
 See The RPA Boom, Air Force Magazine, John A. Tirpak, August 2010 at http://www.airforce-magazine.com/MagazineArchive/Pages/2010/August%202010/0810RPA.aspx.
 See Former ISR Chief Calls for More Autonomy in UAVs, Areas A Technology Blog, by Paul McLeary, Aviation Week, January 25, 2012 at http://www.aviationweek.com/aw/blogs/defense/index.jsp?plckController=Blog&plckScript=blogScript&plckElementId=blogDest&plckBlogPage=BlogViewPost&plckPostId=Blog%3A27ec4a53-dcc8-42d0-bd3a-01329aef79a7Post%3Aca1e0ff0-837f-4b24-8e09-acb9eebe7bb9.
 See MQ-9 to Provide Full HD Video by 2015 after Two-Phased MTS Upgrade, Inside the Air Force, 1/20/12 at http://defensenewsstand.com/Inside-the-Air-Force/Inside-the-Air-Force-01/20/2012/menu-id-290.html.
 See section 220 of PL 106-398.
 Find the website at http://unmannedsystemscaucus.mckeon.house.gov/.
 P. 189 ff., Operation Desert Storm: Evaluation of the Air Campaign, General Accounting Office, June 1997, GAO/NSIAD-97-134 at http://www.gao.gov/archive/1997/ns97134.pdf.
 Pp. 135-139, GAO, Operation Desert Storm: Evaluation of the Air Campaign, at http://www.gao.gov/archive/1997/ns97134.pdf.
 P. 13 of U.S. Unmanned Aerial Systems, Jeremiah Gertler, Congressional Research Service, R42136 at http://www.fas.org/sgp/crs/natsec/R42136.pdf.
 P. ix of Summary of Policy Options for Unmanned Aircraft Systems, Congressional Budget Office, June 2011 at http://www.cbo.gov/ftpdocs/121xx/doc12163/06-08-UAS.pdf.
 There are additional drone costs in the DOD budget, but they are classified and for the Central Intelligence Agency.
 P. vii of Summary of CBO, Policy Options, at http://www.cbo.gov/ftpdocs/121xx/doc12163/06-08-UAS.pdf.
 P. 1, CRS, U.S. Unmanned Aerial Systems,at http://www.fas.org/sgp/crs/natsec/R42136.pdf.
 For a discussion of unmanned systems in the previous century, see Air Force UAVs: The Secret History, Thomas Erhardt, July 2010, A Mitchell Institute Study, at http://www.afa.org/mitchell/reports/MS_UAV_0710.pdf.
 See, for example, the descriptions of Reaper at the General Atomics website at http://www.ga-asi.com/products/aircraft/predator_b.php, which describes endurance as 30 hours without caveat (discussed below in text) and the description by GlobalSecurity.org at http://www.globalsecurity.org/military/systems/aircraft/mq-9.htm, which, for example, describes the performance of Reapers Synthetic Aperture Radar without any issues and available munitions to include GPS-guided Joint Direct Attack Munitions without the issues pointed out by others. The Air Forces factsheet is at http://www.af.mil/information/factsheets/factsheet.asp?id=6405.
 For example, see the Wikipedia entry for Reaper, as of February 2012, at http://en.wikipedia.org/wiki/MQ-9_Reaper.
 See the system descriptions by GAO (at p. 113 of Defense Acquisitions: Assessments of Selected Weapons Programs, Government Accountability Office, March 2011, GAO-11-233SP, at http://www.gao.gov/new.items/d11233sp.pdf ) or by DODs DOT&E at http://www.dote.osd.mil/pub/reports/FY2011/.
 See CRS U.S. Unmanned Aerial Systems. An informative and useful report, it is at http://www.fas.org/sgp/crs/natsec/R42136.pdf.
 See CBOs Policy Options volume at http://www.cbo.gov/ftpdocs/121xx/doc12163/06-08-UAS.pdf. This report contains some basic system information, but is mostly devoted to policy options which are limited to only a mix of current and planned drones, not other platforms that in some cases are clearly cheaper and more effectivesee following text in main body.
 At Cannon and Holloman AFBs, NM; Creech AFB, NV, and Syracuse NY plus several locations in CENTCOM and AFRICOM. The Air Force will start MQ-9 operations at Ellsworth AFB SD in 20012. P. 7 & 11 of United States Air Force, Report to Congressional Committees , Report on Future Unmanned Aerial Systems Training, Operations, and Sustainability, September 2011, submitted to Congress pursuant to House Report 111-491, page 509.
 See the Air Force fact sheet at http://www.af.mil/information/factsheets/factsheet.asp?id=6405.
 P. 13 of CRSs U.S. Unmanned Aerial Systems, at http://www.fas.org/sgp/crs/natsec/R42136.pdf.
 P. 4, Selected Acquisition Report (SAR) RCS: DD-A&T (Q&A) 823-424; MQ-9 UAS Predator, as of June 30, 2010. Not available on line but available from the author.
 Pp. 14-18 of the DOD SAR for Reaper.
 See Service Scales Back MQ-9 Purchases: Lt. Gen. James: Air Force Could Provide More Than 65 CAPs If Needed, Maggie Ybarra, Inside the Air Force, 2/17/12.
 The previous production plan was scheduled to end after 2016; if it is extended to 2017 with the additional 24 air vehicles, the reduction would be 72 rather than 96.
 See the General Atomics advert for Predator B, rather Reaper, at http://www.ga-asi.com/products/aircraft/predator_b.php.
 Find the Air Force Fact sheet on MQ-1B Predator at http://www.af.mil/information/factsheets/factsheet.asp?fsID=122; on Reaper at http://www.af.mil/information/factsheets/factsheet.asp?fsID=6405.
 Hellfire is a short range 100 pound missile with a 20 pound warhead; find a discussion of it at the Wikipedia website at http://en.wikipedia.org/wiki/AGM-114_Hellfire. There is also a useful system description of Hellfire at the Federation of American Scientists website at http://www.fas.org/man/dod-101/sys/missile/agm-114.htm.
 See text in main body below; while the fact sheet asserts Reaper can employ GPS-guided JDAM GBU-38, that appears not to be the case.
 See p. 63 of FY 2009-2034 Unmanned Systems Integrated Roadmap, Department of Defense, 2009, at http://www.acq.osd.mil/psa/docs/UMSIntegratedRoadmap2009.pdf.
 See p. 67 of FY 2009-2034 Unmanned Systems Integrated Roadmap, at http://www.acq.osd.mil/psa/docs/UMSIntegratedRoadmap2009.pdf.
 This data not from USAF Fact Sheet; see instead General Atomics brochure at http://www.ga-asi.com/products/aircraft/predator.php. CBO limits the endurance to 24 hours with no weapons payload; 20 hours with weapons; see p. 4 of http://www.cbo.gov/ftpdocs/121xx/doc12163/06-08-UAS.pdf. A different DOD source limits loiter to 16 hours with external stores; see http://www.acq.osd.mil/psa/docs/UMSIntegratedRoadmap2009.pdf.
 These data not from USAF Fact Sheet; see instead General Atomics brochure at http://www.ga-asi.com/products/aircraft/predator_b.php, and Global Security at http://www.globalsecurity.org/military/systems/aircraft/mq-9.htm. CBO cites 21 and 17 hours for with and without weapons; see p. 4 of http://www.cbo.gov/ftpdocs/121xx/doc12163/06-08-UAS.pdf. DOD cites 24 hours with no external stores at http://www.acq.osd.mil/psa/docs/UMSIntegratedRoadmap2009.pdf.