Monday, July 27, 2015

New GAO Report on CFATS Program

Last Friday the Government Accountability Office (GAO) published their latest report on the Chemical Facility Anti-Terrorism Standards (CFATS) program. While the report did identify some areas where the DHS Infrastructure Security Compliance Division (ISCD) needed to improve the CFATS program it generally noted that significant improvements had been made and previously identified problems had generally been corrected.

The GAO Report identifies four areas of concern:

DHS has not taken steps to mitigate errors in some facility-reported data;
DHS does not have reasonable assurance that it has identified all of the nation’s highest-risk chemical facilities;
DHS cannot ensure consistency in how it addresses noncompliance in the CFATS program because it does not have documented processes and procedures; and
DHS’s CFATS performance measure does not reflect security measures that facilities have implemented and that ISCD has verified.

Top Screen Data Reporting

The GAO identifies a problem with the reporting of Distance of Concern DOC for the release of toxic chemicals in the Top Screen. The CFATS Top Screen requires the facility to calculate the down wind distance that a worse case discharge of a toxic release chemical of interest (COI) will cause a significant problem. The tool that facilities are required to use is the EPA’s RMP*Comp.

The user inputs the maximum amount of a Toxic COI that they have on site, enters some other basic information (see pages 42 and 43 of the Top Screen User’s Manual) and the tool calculates DOC which is then reported in the Top Screen. DHS then uses this information as part of its determination of whether or not a facility may be covered under the CFATS program as a facility at high-risk of terrorist attack.

The GAO used available Top Screen data to verify the DOC reported for a ‘a generalizable sample of facilities’. Using that data the GAO report indicates that 44% of the facilities (2,700 facilities) had errors in the reported DOC and about 43% under-reported the DOC. It goes on to note that a common potential reason for the under-reporting may be due to one difference in the way the tool is used to calculate EPA and CFATS DOC information, the CFATS program does not allow facilities to take credit for passive mitigation measures such as dikes around tank farms.

The Report provides an example of a facility with more than 200,000 lbs of anhydrous ammonia  (AA) reported in its Top Screen that reported a DOC of 0.9 miles and GAO found a minimum possible distance of 2.4 miles when they calculated the DOC using R*Comp. I have replicated that work and found that there was no way to come up with a DOC of 0.9 miles regardless of whether or not mitigation measures were used. I suspect that the facility used their largest storage tank data (as they would for EPA reporting) instead of the total amount of AA on site as required by DHS. It is remotely possible that GAO’s figure of 0.9 miles came from the DOC value reported for the Area of Highest Quantity (AHQ) instead of the total COI.

The GAO report notes that ISCD has all of the information in its Top Screen Database necessary to verify the DOC data, but does not choose to do so. The first part is not necessarily true. The RMP*Comp tool, when calculating the DOC for materials that are gasses at 25°C, asks if the material is liquefied, and if liquefied whether it is liquefied by refrigeration or by pressure. That information is not included in the Top Screen and makes a big difference in the DOC. This is not important for most toxic release COI, but it is for AA. Using the Report’s example with AA you could get DOC’s of 2.4 miles (unliquified) 6.5 miles (liquefied under refrigeration) vs 8.0 miles for liquefied by pressure (all in an urban setting).

The thing that the GAO failed to take into account in pointing out this deficiency is that ISCD does not verify any of the information provided in the Top Screen. It is true that they could generally check the DOC value (using the ‘unliquified’ data from RMP*Comp), but that might not give a true picture for all COI. But given the fact that ISCD is accepting all other reported information, it would be unusual for them to pick out this one item that could be partially verified in a portion of the instances where it is reported.

Given the fact that the Report notes that only 43% of the discrepancies that it noted were under-reports, it seems to me that their data would tend to indicate that there were systemic problems with the use of the RMP*Comp tool. As ISCD moves forward with implementing the results of the outside evaluation of their risk ranking methodology, they should consider taking this calculation out of the hands of the facility and do the calculations in-house.

The GAO has two recommendations for this area:

Provide milestone dates and a timeline for implementation of the new Top-Screen and ensure that changes to this Top-Screen mitigate errors in the Distance of Concern submitted by facilities, and
In the interim, identify potentially miscategorized facilities with the potential to cause the greatest harm and verify the Distance of Concern these facilities report is accurate.

Facility Identification

The report outlines the measures that DHS has taken to identify facilities that have not submitted Top Screens, but should have done so. While they had conducted earlier out-reach activities, the effort was expanded after the West Fertilizer incident and the issuance of the President’s Executive Order on Increasing Chemical Facility Safety and Security. As a result of these latest efforts just over 3,000 potentially non-compliant facilities were identified and contacted by DHS.

More than 1500 had already submitted Top Screens; the ‘new’ identification was apparently based on differing naming or location information. Over three hundred were exempted from CFATS regulations. Of the remaining just over 1,000 have now submitted Top Screens and just 24 of those have been designated as high-risk facilities covered under the CFATS program with 44 still pending. ISCD is continuing to investigate other means of identifying potentially non-compliant facilities.

The report indicates an interesting problem. ISCD has asked States for information on the chemical facilities that they regulate as part of this program. California recently complied, identifying over 46,000 facilities (ISCD has only processed 50,000 Top Screens since the program started) which ISCD is now going through. Only 13 other States have supplied similar lists.

The GPO did not provide any recommendations for DHS on this issue.

Compliance Inspection Issues

The Report starts of the discussion of this issue with a review of actions that ISCD has taken to increase their rate of site security plan approvals and notes that ISCD has made substantial improvements in that approval rate. Interestingly, even though the GAO calculated that ISCD would have the approval backlog eliminated next year, they did not mention that the EAP process will almost certainly further accelerate the SSP approval process.

The Report then notes that ISCD has completed 83 compliance inspections of facilities with approved site security plans. There is no discussion of how well that reflects the requirement for ISCD to inspect facilities within one year of their site security plan being approved. The number seems low, but it will almost certainly increase as ISCD has fewer authorization inspections to complete.

The GAO reports that nearly half of the facilities inspected have not completely implemented all of the security measures outlined in their site security plans, which of course means that the facilities are out of compliance. The Report notes that ISCD is working with the facilities to get them into compliance.

The GAO notes that none of the non-compliance sanctions available to the Department (including Compliance Orders, Civil fines and even Cease Operations Orders) have been used by the Department to-date. While ISCD is ‘working with’ the facilities, the GAO reports that they do not have any written processes or procedures in place to document the progress that is being made at those facilities. Nor, apparently, does ISCD have any written processes or procedures in place on how they determine whether or not a facility is in compliance.

The GAO had one recommendation for this area:

Develop documented processes and procedures to track noncompliant facilities and ensure they implement planned measures as outlined in their approved site security plans.

CFATS Performance

The last area of concern identified in the Report concerns the program reporting done by ISCD to DHS. This annual reporting requirement is used by DHS and the GAO to assess program performance and efficacy. One of the pieces of information included in that report is the number of security measures implemented by facilities. The GAO notes that ISCD does not distinguish between those measures implemented before the facility site security plans were approved, which measures have been reported as planned, or which of the planned measures have been implemented. Thus, the GAO reports that the numbers do not reflect changes brought about by the CFATS program and ISCD actions in support of that program.

The GAO had one recommendation for this area:

Improve the measurement and reporting of the CFATS program performance by developing a performance measure that includes only planned measures that have been implemented and verified.

Moving Forward

DHS has acknowledged the four recommendations in the GAO report and has reported their intended actions to be taken in response to those recommendations. GAO confirms that if those actions are taken as reported, the recommendations would be considered as completed.

No comments:

/* Use this with templates/template-twocol.html */