HomeFeaturesDailyBriefingsRapidReconSpecial ReportsAbout Us

Combating Nuclear Smuggling

A GAO (Government Accounting Ofice) report released last week found that the Department of Homeland Security's Cost-Benefit Analysis used to determine contract awards was not based on the performance data of the tests, and did not consider all of the costs and benefits of the systems.

Since September 11th, the U.S. government has spent $350 million to install more than 840 radiation detection devices at borders, seaports, and international mail centers.

The government contracted with foreign operators such as Hong Kong-based Hutchison Whampoa to operate radiation detection machinery in foreign ports (notably Freetown in the Bahamas). Most of that detection machinery relied on crude detection technology used in scrap-metal recycling. Although this equipment was sufficient for industrial use (without the ability to identify the exact isotopic source of the radiation), for security purposes the equipment was subject to numerous false positives with modestly radioactive but harmless substances such as cat litter, ceramic tile, and bananas setting off alarms. To avoid the resulting delays caused by these false alarms, port officials recalibrated the detection sensitivity of the devices. While reducing the false alarm rate, this made the detection devices less capable of detecting threatening sources of nuclear radiation.

The new technology, which DHS calls Advance Spectroscopic Portal monitor, simultaneously detects the presence and type of radiation. The department has already awarded contracts totaling $1.16 billion to Meriden, Connecticut-based Canberra Industries; Waltham, Massachusetts-based Raytheon; and Waltham, Massachusetts-based Thermo Electron, to develop the new technology. DHS officials say that the accuracy of the new technology will reduce the number of secondary inspections to about 15,000 a year. There is one drawback to the new technology: Cost. The new machines will cost about $500,000 each, which is the price of seven of the old detection machines. Still, current DHS plans call for purchasing 1,400 of the new machines by 2011.

About a month ago, the Department of Homeland Security had awarded contracts under two related programs.

Advance Spectroscopic Portal - In July DHS awarded $1.16 billion worth of ASP contracts to three vendors -- Raytheon, Thermo Electron, and Canberra -- for one base year plus four annual options. The goal for the first year is to develop a fixed radiation detection portal which will become the "standard installation for screening cargo containers and truck traffic." Here is the news release from DHS announcing the award.
Cargo Advanced Automated Radiography System - The Department of Homeland Security awarded contracts totaling $1.35 billion to L-3 Communications Holdings Inc., American Science & Engineering Inc. and employee-owned SAIC to screen for radioactive material in hidden cargo. The DHS program is called the Cargo Advanced Automated Radiography System, or CAARS, and it is designed to deliver "an advanced imaging system that will automatically detect high density shielding that could be used to hide special nuclear material such as highly enriched uranium or weapons grade plutonium."

This all seems pretty straightforward. DHS requests proposals for a system or systems that meet critical national requirements. But last week, the GAO (Government Accounting Office) published a report titled: DHS's Cost-Benefit Analysis to Support the Purchase of New Radiation Detection Portal Monitors Was Not Based on Available Performance Data and Did Not Fully Evaluate All the Monitors' Costs and Benefits.

The lives and safety of Americans are at stake here. And yet, according to the GAO report, the DHS and its newly formed Domestic Nuclear Detection Office (DNDO) failed to follow its own testing report and protocols. THe GAO report revealed that:

1) tests of ASPs showed that they did not meet DNDO’s main performance assumption in the cost-benefit analysis of correctly identifying unmasked highly enriched uranium (HEU) 95% of the time it passes through portal monitors (actual performance was in the 70-88% range);

2) identifying masked HEU, considered much more difficult, was detected by the three winning companies 53%, 45% and 17% of the time respectively (the GAO report did not match the detection percentages with the names of the three companies);

3) the DNDO, while acknowledging that none of the ASPs met the standards in the independent testing, assumed that the units would eventually reach that level of performance sometime in the future;

4) the DNDO’s cost-benefit analysis only considered the ability of the ASPs to detect masked and unmasked HEU, and not any other type or form of radiological material (apparently, this capability is dependent on whether the ASP has the software to perform the detection of these other materials);

5) the DNDO apparently did not follow DHS guidelines for performing the cost-benefit analysis, omitting such variables as the difference in cost between land based cargo portal monitors and seaport units, despite acknowledging that there are as many as 12 different ASPs;

6) life-cycle costs for operations and maintenance for the equipment over time was underestimated;

7) the DNDO did not assess the likelihood that radiation detection equipment would either misidentify or fail to detect nuclear and radiological materials. Rather, DNDO’s cost-benefit analysis focuses on the ability of ASPs to reduce false alarms. This leads to the possibility (discussed in the GAO report) that the ASP could misidentify HEU and allow it to pass through a portal.

The entire report includes a brief Powerpoint presentation, as well as DHS/DNDO explanations for the variances from the cost-benefit analysis guidelines. However, considering the importance of port security, preventing any sort of nuclear or radiological weapon being detonated, and the significant amount of money involved in the program, we should all be concerned.

Do these new and expensive radiological detectors work as specified and expected? Will they work under operational conditions? And perhaps more importantly, what influences caused the DMDO to not follow the guidelines for the cost benefit analysis?

Notes