Big Money, Fences, and Drones: An Excerpt from Border Insecurity

Big Money, Fences, and Drones: An Excerpt from Border Insecurity

The following is an excerpt from Breitbart Texas’ border security expert and contributing editor Sylvia Longmire’s new book, Border Insecurity: Why Big Money, Fences, and Drones Aren’t Making Us Safer:

Since the beginning of our government’s efforts to make our southwest border less permeable, there has never been any intention to completely fence the entire two thousand-mile length of it–at least, not with a physical fence. After all, the Secure Fence Act and subsequent legislation only called for seven hundred miles of it in the most highly trafficked areas. But DHS felt it was important to have some sort of surveillance coverage along the unfenced areas. It needed to find a way to achieve this that was cost effective and quick. However, the agency was also painfully aware that it did not possess the in-house expertise to accomplish this, so the outsourcing of our physical border security began.

With the announcement of the SBI strategy in 2006 came the introduction of SBInet–a program that, in theory, would bring together the brightest minds and best border technology available to create a network of surveillance platforms along the unfenced parts of the border. The program was soon commonly known as the virtual fence, and was doomed to become the most colossal failure in the history of US border security efforts. Even more damning: such a project had been tried twice before, with very discouraging results.

An earlier attempt at mixing technology with traditional fencing came in 1998, when the former Immigration and Naturalization Service announced a system–the Integrated Surveillance Intelligence System (ISIS)–that would connect new ground sensors, towers with cameras, and computers to identify and interdict illegal border crossers. The initial contract was for the modest sum of $2 million and was awarded without the usual competitive bidding process to International Microwave Corporation. Walt Drabik, the ISIS project director, claimed in 2000 that the thirteen thousand old unattended ground sensors would be replaced with ones that were more sensitive and less likely to result in false alarms. They would then be connected to seventy-three high-resolution and infrared cameras in twenty-one Border Patrol sectors. 

In fact, this sensor integration should have happened a decade earlier but never materialized. That 1989 project was called the Intelligent Computer-Aided Detection system, or ICAD. Information gleaned from the integrated sensors was supposed to be analyzed by law enforcement communications assistants, who would then filter the most important information to Border Patrol agents who could respond to a sensor alarm. ICAD went through a series of modifications, or “improvements,” in 1994 and sometime later entered its third iteration. ISIS would come to incorporate the data gathered through this system.

The biggest problem with ICAD and ISIS was that the data they were basing their systems on–coming from thousands of bad sensors–was fundamentally flawed. Over the course of eleven years, no one seemed to have caught the fact that everything they were collecting and ultimately producing was garbage because it was based on false alarms or unknown causes of sensor trips. However, that didn’t stop ISIS from soldiering on; in 2000, Congress allocated $200 million to keep developing. Two years later, after audits, it became apparent that the $239 million appropriated for the virtual security project known as ISIS had been completely spent by contractors on a system that was “functionally inoperable.”

It isn’t clear why DHS chose to pursue the construction of another technological border barrier so soon after the ISIS debacle, or what members of Congress who approved funding for SBInet were thinking when they decided to slap down $1 billion for its construction. Regardless, Boeing was awarded the contract to build the new virtual fence after claiming it could complete the first twenty-eight miles of it–later called Project 28 in Arizona–in just eight months, and completely wire the southwest border in three years. Boeing also claimed that its system would be able to identify and initiate apprehension of approximately 95 percent of all border crossers. Needless to say, things didn’t exactly work out as planned.

Instead of creating a customized system of sensors and cameras that were optimized for operation in the various southwest border environments, Boeing just grabbed whatever gadgets it happened to already have from other projects (called COTS for “commercial off-the-shelf”) and tried to tweak them for this purpose. The upside of this tactic is that the systems were familiar and had been tested for functionality. The downside of this is that Boeing could charge whatever DHS was willing to pay them without having to invest that money in research and development or any other overhead costs associated with creating a custom-built system. Never mind the fact that no one bothered to ask Border Patrol agents what they wanted or needed from the virtual fence, leading Boeing to integrate components that agents later found cumbersome at best and useless at worst.

From the start of SBInet, some members of Congress expressed concern about the amount of money involved in the project, as well as the lack of a concrete plan for moving forward with construction. In early 2006, before the virtual fence contract was awarded to Boeing, Homeland Security Appropriations Subcommittee Chairman Harold Rogers (R-Kentucky) told Homeland Security officials during a hearing, “We’ve been at this juncture before. We have been presented with expensive proposals for elaborate border technology that have eventually proven to be ineffective and wasteful systems, such as the Integrated Surveillance Intelligence System and America’s Shield Initiative . . . When presented with questions like this, we apply a simple formula: no plan equals no money.” During the same hearing, Homeland Security Appropriations Subcommittee ranking member Martin Olav Sabo (D-Minnesota) added, “I’m worried that DHS thinks that the solution is to hire a private technology company to run the SBI, and then sit back and watch.”

And that’s exactly what happened. But DHS didn’t dive into the virtual fence head first; it decided to get its feet wet with Project 28,which was plagued from the start with cost overruns, delays, and not enough CBP employees to run the system. The CBP employees DHS did have on hand to run Project 28 had little to no project management experience, resulting in fiscal mismanagement and an inability to hold Boeing accountable for any contract missteps. Still, Project 28 continued with the goal of erecting nine towers upon which state-of-the-art surveillance systems would be mounted. Because the systems were “off-the-shelf,” many at both Boeing and DHS thought the cameras and sensors would just snap into place and start working with little fuss; this accounts for the quick turnaround time projected by Boeing management. 

Then the setbacks began. Boeing engineers couldn’t work out bugs in the system. Rain and blowing sand rendered the surface radar–technology that’s been in use for more than half a century– on the towers useless. The type of radar integrated into the towers was designed to detect large military vehicles, and was less than stellar at identifying smaller objects like people. The computer system also couldn’t be integrated properly. DHS didn’t have enough people with the skills or training to use the finished product.

In 2008, then-DHS Secretary Michael Chertoff announced the completion date of the virtual fence would be pushed back three years to 2011. The SBInet budget by mid-2009 had reached $770 million, and a new government stimulus package provided an additional $200 million in funding. At around this time, both CBP and Boeing management acknowledged Project 28 was a mess. Engineers admitted they cobbled together components, some of which weren’t even theirs and were purchased from Radio Shack and other retailers. All this came out after CBP accepted Project 28 as functional in 2008. And despite this, plans continued to roll out the next iteration of Project 28, known as Tucson-1 and Ajo-1 and hailed as the “real” SBInet.

The new project turned out to be as much of a dud as its predecessor. Tucson-1 and Ajo-1 comprised fifty-three miles of southwest border in Arizona, and to this day is the only section of the US-Mexico border to be touched by the virtual fence. This is because in January 2010, former DHS Secretary Janet Napolitano called for a systemic review of SBInet. Specifically, she wanted to know if the virtual fence was viable, if it could meet the original goals (which were never clear in the first place), and whether it would be cost effective in the long run.

In October 2010, the GAO issued a report titled, “Secure Border Initiative: DHS Needs to Strengthen Management and Oversight of Its Prime Contractor.” Yes, that really was the title of the report. In January 2011, after spending $1 billion of taxpayer dollars on only fifty-three miles of virtual fence that mostly didn’t work, Napolitano announced that SBInet systems “were not appropriate for the entire southwest border and did not meet current standards for viability and cost effectiveness.” The initial cost estimate that Boeing provided in 2006 for the virtual fence, which would consist of 1,800 towers once completed, was $2 billion. That estimate was later raised to $8 billion, and the Inspector General suggested the true cost down the road would be closer to $30 billion. The original completion date for the full virtual fence was 2009, and that was ultimately pushed back to 2013.

Still, DHS continues its trial-and-error strategy for finding a technological solution to securing our borders. Again, Arizona will be the test bed for the technology acquisitions plan, dubbed Alternative (Southwest) Border Technology. It calls for combining the existing virtual fence infrastructure with integrated fixed-tower systems (IFTs), remote video and mobile surveillance systems, unidentified handheld equipment, and unattended ground sensors (no indication if they mean existing sensors or new ones that actually work). The projected life cycle cost for CBP’s plan is $1.5 billion.

You can read more about the good the bad, and the ugly of border technology, as well as Washington’s inability to define what a secure border looks like in order to develop a comprehensive border security strategy, in Sylvia Longmire’s book Border Insecurity. You can learn more about the book, watch the official trailer, and read reviews at the official website,