Category Archives: NGA

NGA’s “Map of the World” not exactly open

The National Geospatial-Intelligence Agency (NGA), America’s biggest geography spy agency, has posted an invitation on its media PR page to hear about its “Map of the World” project.

If you’d like to know more about this, the NGA starts off by noting:

Collaboration with industry and academia is critical to the NGA mission, our customers and our strategic partnerships.

Thinking of signing up to learn more? Maybe not:

Except for a one-hour unclassified information session, this event is limited to U.S. citizens holding top secret clearances.

Remember we pay the NGA at least $4.9 billion per year. It seems “industry” and “academia” can only take part by becoming part of the intelligence community.

Our new paper on intelligence now online


Very excited to announce our paper “The New Political Economy of Geographical Intelligence” is now online at the publisher’s website for the Annals of the Association of American Geographers.

The publishers have provided a link for free access to the first 50 people (click here for free access)! (Edit: these have unfortunately all been claimed)

The regular link which will remain after those free accesses are used up is this one


A troubling new political economy of geographical intelligence has emerged in the United States over the last two decades. The contours of this new political economy are difficult to identify due to official policies keeping much relevant information secret. The U.S. intelligence community increasingly relies on private corporations, working as contractors, to undertake intelligence work, including geographical intelligence (formally known as GEOINT). In this article we first describe the geography intelligence “contracting nexus” consisting of tens of thousands of companies (including those in the geographical information systems and mapping sector), universities and nonprofits receiving Department of Defense and intelligence agency funding. Second, we discuss the “knowledge nexus” to conceptualize how geographical knowledge figures in current U.S. intelligence efforts, themselves part of the U.S. war on terror and counterinsurgency (COIN). To analyze the contracting nexus we compiled and examined extensive data on military and intelligence contracts, especially those contracts awarded by the country’s premier geographical intelligence agency, the National Geospatial-Intelligence Agency (NGA), for satellite data. To analyze the knowledge nexus we examined recent changes in the type of geographical knowledges enrolled in and produced by the U.S. intelligence community. We note a shift from an emphasis on
areal and cultural expertise to a focus on calculative predictive spatial analysis in geographical intelligence. Due to a lack of public oversight and accountability, the new political economy of geographical intelligence is not
easy to research, yet there are reasons to be troubled by it and the violent surveillant state it supports.

Key Words:
geographical intelligence, geographical knowledge, GEOINT, government contracting, National Geospatial-Intelligence Agency.

Annals paper on Geographical Intelligence

I’m very excited to say that our (myself, Sue Roberts and Ate Poorthuis) paper for the Annals of the Association of American Geographers is now at the proofs stage. The first page is below. I believe it will be out in an early 2014 issue.

Pages from Proofs

PDF of first page

Harvard CGA presentation

Last week I was a panelist at the annual Center for Geographic Analysis conference at Harvard University. The panel topic was on privacy and legal aspects of locative media and the geoweb.

My remarks focused on what I’m coming to call “sustainable security” and posited that in many respects we have lost personal privacy and are willing to lose even more for the promise of being safer. Sustainable security seeks a level of security that is sustainable in the long run, and does not accept that all security is good or that it makes us safer.

Are we safer for all the money we spend on security? I suggested three reasons why we might not be (I’ve pasted my ppt below and I believe it was recorded on video).

1. outsourcing is growing the old military-industrial-academic sector, this outsourcing is for-profit, and leads to lack of oversight;

2. geographical research is being enrolled into the securitization agenda, converting “peaceful” research into militaristic purposes;

3. numerous legal provisions which enable the security state undermine civil liberties (not just privacy).

My co-panelists were Kirk Goldsberry (Harvard CGA), David DiBiase (Esri) (my first boss at Penn State) Nicolas Oreskovic, and Sarah Williams (MIT). We might have had some good discussion but unfortunately one of the panelists spoke for at least double the allotted time (we were asked to speak for about 10 minutes).

I would have especially liked to engage more with David, who felt I was being hypocritical in flagging Esri’s government (and DoD) contracting without mentioning the huge sums pulled in by universities (including UKY). This is certainly true–universities like UKY, and George Mason, Penn State and indeed Harvard do take large government contracts (although I was told that Harvard does not allow “secret” research)–but I think the reason to flag Esri is that (along with DigitalGlobe and GeoEye) it is a top ten contractor with the NGA, and as a privately held company does not publish its contracting details with DoD and NGA. (For its part, the NGA has had a special dispensation since 2006 not to report dollar figures it contracts out.) Esri is also a major mover in the GIS/locative space, and government funding of its efforts, and its role in working with DOD combat-support intelligence agencies, is worthy of note.

Nevertheless, as I noted in a tweet reply to David, he is right to wish for a more complete picture of the political economy of the geoweb. This is something I’d like to work on a bit more, and something some colleagues and I have discussed in more length in a paper in submission. 

Data on government spending is available at a couple of different places, eg., and, but for both the source is the Federal Procurement Data System (FPDS). According to the FPDS the following government awards have been made:

  • Esri $878m ($479m from DoD)
  • University of Kentucky $260m ($23m from DoD; including four awards over $1m, all from the Army)
  • George Mason University (a previous employer), $185m ($78m from DoD)
  • Penn State (David’s previous employer and my alma mater) $2.8b ($2.5b from DoD)
  • Johns Hopkins (top university recipient of federal contracts) $9.7b ($9.1b from DoD)
  • Harvard University $146m ($29m)

Penn State and Johns Hopkins’ size seriously dwarf the other examples given here. Harvard’s relatively tiny numbers (for an institution with a $32b endowment) are presumably a factor of its refusal to do “secret” research.

It would be worth looking at what these institutions are being funded to do, but here we run into problems of accessibility, absent sending in multiple FOIA requests. (If Esri wants to release financial summaries of its government and DoD contracts, I would be obliged, and give it due credit for transparency.)

These flows of money may seem like a lot, but in context of government and DoD funding they are relatively small. (Perennial government consultants Booz Allen Hamilton has absorbed more than $11.3b of tax-payer’s money; the top government contractor of all time is Lockheed Martin at $393b–how’s that for corporate welfare?)

But surely dollars are only part of the story. There’s also personnel and the rotation of government contractors, such that contractors become government  employees with the power to award contracts (and vice versa). This is partly why so many contracts are awarded on an “uncompeted” or if competed then only done so by a single company.

As I allude to in point 2. above, there are also the circulations of knowledge between various parts of the academic-military-intelligence-contractor “nexus.”  These are harder to pin down but potentially more illuminating. In my case, I am interested in how geography gets put to work.

In other words, how does expertise (concepts and methods) get co-opted (advertently, inadvertently) and circulate?

Take this story in The New Republic for instance. They found

at least 49 people who have simultaneously worked as lobbyists for outside entities while serving as top staff, directors or trustees of 20 of the 25 most influential think tanks in the United States.”

The list includes people who worked on national security issues at think tanks, but simultaneously lobbied for military munitions manufacturers:

[Center for American Progress] senior fellow Scott Lilly, whose beat includes national security issues, lobbied for Lockheed between 2005 and late-2011.

Should this reduce the credibility of Lilly or CAP? What if the financial support gained is not disclosed or only partially disclosed? Would transparency completely fix the problem–is it enough?

We also have this:

This should remind us that there is no such thing as the efficacy of pure knowledge–knowledge has a politics. How can we then develop reputable or consensual knowledge about what constitutes sustainable security (rather than security for security’s sake)? And of what do those consensual knowledges consist?

Harvard CGA slides: Sustainable Security

GeoIQ, Esri and NGA

The National Geospatial-Intelligence Agency (NGA), which is a DoD combat-support mapping and geospatial imagery intelligence agency, recently provided a short update (pdf) on GeoIQ, which was bought by Esri last year. GeoIQ has also received funding from the CIA’s venture-capital company, In-Q-Tel.

Even in an environment bursting with practitioners,  many at the National Geospatial-Intelligence Agency recognize the need for  geographical analysis tools that can be accessed and enhanced by users across the customer base. 

GeoIQ was one platform that particularly stood out to NGA and In-Q-Tel,  because it allows users to collect massive amounts of information and analyze it through geospatial analysis, said Jay Brennan, deputy program manager for the NGA IQT Program.

“I think a product like GeoIQ fits nicely with the (NGA) director’s desire to get GEOINT into the hands of users,” said Brennan. “It does a good job of making it easier for the non- (Geographic Information System) person to create their own maps, bring in their own data and do some first-level analysis. This frees up NGA analysts to do second-order or third-order analyses.”

In-Q-Tel is a private, independent organization that identifies and partners with companies that produce technologies of interest to the U.S. intelligence community. Since 2003, In-Q-Tel has received more than $17 million in funding from NGA. On average, for every $1 NGA spends, In-Q-Tel leverages an additional $37 from venture capital firms and the larger U.S. intelligence community, allowing NGA to maximize the impact of its investment.
In-Q-Tel made its first investment in GeoIQ in May 2007. Since then, the technology has found a place in the Integrated Work Group-Readiness, Response and Recovery (IWG-R3), where it is used within NGA and by many of its outside customers, such as first responders in the field.

GeoIQ has been employed during the recovery efforts following the 2010 earthquake in Haiti, in the wake of hurricanes, and after other natural disasters, said Nathaniel Wolpert, an IWG-R3 geospatial intelligence analyst.

IWG-R3 uses GeoIQ to geospatially visualize and determine areas that are flooded or in need of supplies, and to understand the overall situation on the ground, said Wolpert.

NGA maintains a contract with GeoIQ, which was acquired by California-based Esri in July 2012, to source additional capabilities and tools based on the technology. NGA has provided feedback throughout its development and the company has been and continues to be very responsive, said Wolpert.
“It’s a powerful tool that we want to keep using and make more robust in the future,” said Wolpert.

Nearly 5 million hold security clearances

The number of people holding security clearances rose to about 4.9m in 2012, according to the latest official figures (pdf) from the Office of the Director of National Intelligence (ODNI).

Steven Aftergood of the FAS Security blog provides more details:

The total number of cleared personnel as of October 1, 2012 was 4,917,751.  Although the number of contractors who held a clearance declined in 2012, the number of eligible government employees grew at a faster rate, yielding a net increase of 54,199 clearances, or 1.1 percent, from the year before.

It is possible that there were more security-cleared Americans at some points during the Cold War, when there was a larger standing military with more cleared military personnel than there are today.  But until 2010, no comprehensive account of the size of the security clearance system had ever been produced.  So the new 4.9 million figure is the largest official figure ever published.

As he notes, ODNI requested that the legal obligation to report these numbers be canceled, which was initially granted, but following a public outcry, the obligation was restored (ODNI claimed it took a lot of time and effort to prepare).

It’s very useful for us to know these numbers, and indeed further information could well be disclosed without threat to national security, such as the number of contractors working at each IC agency, and in what capacity.

Spy satellite imagery to be released

Steven Aftergood of the Federation of American Scientists (FAS) has an update on the release of KH-9 satellite imagery to the public:

Intelligence community officials have been meeting with representatives of the National Archives to discuss the anticipated declassification and release of intelligence imagery from the KH-9 satellite dating between 1971 and 1984.

Officials have been negotiating the transfer of the original negatives from the KH-9 system and the provision of finding aids, according to a newly released but heavily redacted report from the National Geospatial Intelligence Agency, dated June 2012.

If “keyhole” reminds you of keyhole markup language (kml) it should. Keyhole was the name of the software which was bought by Google and turned into Google Earth. I interviewed Avi Bar-Zeev, one of the co-founders of the company in Cartographica (article is freely available here).

The June 2012 report was obtained by Aftergood through FOIA. As he correctly points out, it is quite heavily redacted, but it details the release of historical imagery from the KH-9 systems, operational between 1971 and 1984 (and officially decommissioned in 2011).

How much does the US spend on “security”?

How much does the US spend on security each year? Taking everything into account, and including the Pentagon military base budget and overseas contingency operations (foreign wars), intelligence, Department of State, past wars that have to be paid off, etc, the total comes out to $1.2 trillion, according to Chris Hellman, of the National Priorities Project.

That’s a good base number to keep in mind during talk of sequestration (obligatory spending cuts if no budgets are agreed to).

How does this affect spending on military/intelligence contractors? The pattern is clear: since 2006 contractors have shrunk the number of people they employ (jobs they provide) while at the same time increasing the amount of federal contracting dollars they take in.

As David Swanson puts it, “the logic of bigger contracts = more jobs is essentially a bucket of hope and change.”

What remains to be seen is if smaller contracts = fewer jobs. But here’s two indications it won’t necessarily be so simple.

NGA Watch: Commercial and mobile apps

The NGA is holding an “Industry Day” for commercial and mobile app developers.

According to the NGA there will be a “compelling discussion on NGA’s emerging app (mobile / web / widget) and GEOINT App Store ecosystem as well as associated compensation models.”

Here are the suggested attendees:

Chief Technical Officers, Senior Technical Engineers, Mobile / Web / Widget Analysts & Developers, Contracting Officers, Company Business Development Representatives, Business / Program / Project / Production Managers, and Sales & Marketing Representatives.

Event will be September 27, 2012 and is unclassified.

NGA hires GeoIQ for social media analysis

The National Geospatial-Intelligence Agency has announced a request for quote (RFQ) from FortiusOne/GeoIQ–who operate the popular GeoCommons web-based mapping site–to perform social media datamining in the event of disasters (either man-made or natural).

NGA announced on FedBizOpps that GeoIQ is the only company to be approached (sole-source without competitors) as it is a follow-on to a previous contract held by GeoIQ. It will be what is known as Firm Fixed Price–Time and Materials, according to FAR rules, the government’s primary contracting guidance book.

FortiusOne/GeoIQ/GeoCommons is run by CEO is Frank Moyer, founder and president Sean Gorman, and CTO Andrew Turner. However, this summer Esri bought out GeoIQ for an undisclosed sum.

It is unclear if Esri knew about this contract already and bought GeoIQ because they hoped to expand this line of work. Both Esri and Gorman have confirmed that “existing” GeoIQ customers will continue to be supported, presumably including the NGA. Gorman did reveal that they have an ongoing R&D project called “Twitch” which “handles dynamic aggregation and visualization of millions of points from social media streams with in-browser HTML5 support.” (Added: video demo of the Twitch project).

The statement of work (SOW) for NGA contains similar capabilities:

1.1 Create Temporal Histogram of tweet
1.2 Create predefined hashtags searches per disaster type with ability to customize (languages, abbreviations mispellings etc)

Summarize hashtag searches
o per USNG 1k cell
o per nearest cell phone tower
o per unique user
o Be able to locate last tweet of specific user
o Per user defined areas smaller then 1k (USNG 1k functionality at tighter scale)
o Per tweet within XXX (User defined) distance of predefined infrastructure datasets
o Create raster product of tweets per USNG cell and normalize by Day & Night Population
1.3 “Crawl” tweets and apply hashtag searches to basic text embedded in the tweet and not just hashtag

Task 3 also requires GeoIQ “tie-in” to Facebook, YouTube and “other platforms.” It is not clear however if this is the Twitch project or a separate one.

An attachment provided further details on the SOW, confirming for example that GeoIQ underwent “prior development” by In-Q-Tel (the CIA venture capital company) in 2007. (See press release here).