30 Day CFATS PSP ICR – Moving Forward

This is part of an ongoing series of blog posts about the recently published 30-day information collection request (ICR) published in the Federal Register by DHS. This ICR would support the long overdue personnel surety program requirements for the Chemical Facility Anti-Terrorism Standards (CFATS) program. Earlier posts in the series include:

30 Day CFATS PSP ICR Published

30 Day CFATS PSP ICR – Three Options

30 Day CFATS PSP ICR – Positive Match

With less than 2 weeks left in the comment period, and no comments posted to the www.Regulations.gov web site it may be time to look at what this PSP would look like in actual practice. First, we need to remember that ISCD is only going to apply the PSP requirements to Tier 1 and Tier 2 facilities. This was done to reduce the initial work load on the new PSP system and give ISCD a chance to work the bugs out before they apply it to the bulk of the facilities. This means that another round of ICRs will be required to make that change since the current burden estimate is only based upon the participation of the Tier 1 and Tier 2 facilities.

I also understand that ICS is going to go back to their earlier rollout method of initially only requiring a limited number of Tier 1 facilities to implement the PSP. This will allow those facilities to have their Chemical Facility Inspectors (CSI) on hand during the start up to help work through any of the problems in the system. This was successfully used in the initial rollout of the Top Screen and Security Vulnerability Assessment tools.

System Design

There is going to be some time lag between the time that OMB approves the ICR and the actual implementation of the PSP tools in CSAT. This is because DHS has spent only a limited amount of time and money on developing the tools and manuals. Given the history of this program, I think that we can forgive the ISCD team for thinking that they might be required to make some changes in their current plan by the time OMB gets done with their approval process.

I think that we will see a delay of at least 60 to 90 days between the time that OMB approves the ICR and the time that ISCD announces the initial deployment of the CSAT tools and the limited initial roll out.

Registration Tool

One thing that is going to have to change is the current CSAT Registration Tool. Currently the facility registers specific people to allow them to have access to the various portions of CSAT that affects that facility. Currently the tool allows for the registration of an Authorizer (Executive responsible for CFATS implementation at the facility), Submitter (Person who actually submits completed information to ISCD via CSAT), Lead Preparer/Preparers (the folks that actually enter data into the various tools) and Reviewers (people that are authorized to look at but not touch CSAT information).

For facilities that are doing all of their own data submission in-house, there will probably be a need to add one or two folks from HR to the list of Preparers for the facility. This will not require any CSAT changes.

For facilities that are going to rely on an outside agency to handle the submission of data for their PSP, things get a bit more complicated. The easy way out (and as usual the worst way to do things) would be to authorize one person at the background check agency (BGCA) to do all of the submissions for the facility; this could be done under the current registration rules. The reason that this is the worst way to handle the registration is that we all know that there will not be just one person handling all of the data submission from the BGCA. With just one person ‘registered’ there will inevitably be login credential sharing which tends to compromise the security of the system, a system that will be handling Personally Identifiable Information (PII).

What I suspect that ISCD will do will be to allow a facility to register the use of a BGCA. The BGCA will be enrolled in the ISCD PSP and will register individual employees as Preparers for the BGCA. This will make things simpler for everybody involved. This will also allow vendors and contractors to provide information to a BGCA so that their employees that require access to CFATS facilities on a routine basis can be easily vetted for multiple facilities.

PSP Tool

With the use of BGCA I suspect that we will see effectively a dual PSP tool; one for facilities and one for BGCAs. I think that it may be listed as a single tool, but depending on how one signs in you will see two different sets tools. The basic data being submitted will be exactly the same set of PII, but there will have to be some way for the BGCA to indicate for which facility that PII will be submitted.

I would like to make a suggestion here. I think that it would be much simpler (and eliminate a number of potential errors). The BGCA should be allowed to enter an individual’s PII into the ISCD PSP tool without a chemical facility initially being listed. As they were notified by their clients (vendors, contractors and potentially even individuals) that a person was going to need to have access to a facility, they would add a facility identification number to that individual’s PSP information. Since that person would already be vetted through the PSP, the 48 hour notice would not be necessary and ISCD could send a message to the facility that the person had been vetted through the PSP.

How Long?

On March 5^th, barring some unforeseen eventuality, the folks at NPPD will submit this PSP ICR request to the OMB’s Office of Information and Regulatory Affairs (OIRA). The big question is how long the approval process will take at OIRA. I have seen ICRs approved on the day of their submission, but those were either entirely non-controversial simple exercises or they were politically driven by the Administration. Neither of those applies to the CFATS PSP ICR.

A large part of the inevitable delay in OIRA is trying to work out the political bugs in the program. The more people (or the more powerful the people) that complain about an ICR the longer it will take.

I expect that we will see some negative comments from the same people that complained about the 60-day ICR. Some will go through the eRulemaking Portal, but most will go directly to OIRA outside of public scrutiny. Many of those will be politely ignored and OIRA will try to iron out compromise solutions with the complainer and NPPD/ISCD. How long that will take is anybody’s guess.

I will be very surprised if it takes less than 60 days and I would not be very surprised if it takes six month. The longer it takes past six months, however, the more likely it will be that NPPD will again have to withdraw the ICR and start all over again. I give that about a 40% chance of occurring.

STB Announces Meeting of RETAC, 3-6-14

Today the Surface Transportation Board (STB) published a meeting notice in the Federal Register (79 FR 9587-9588) for a meeting of the Rail Energy Transportation Advisory Board on March 6^th, 2014. Potential agenda items for this meeting include:

• Introduction of new members;

• A performance measures review;

• Discussion of domestic oil production and transportation;

• Industry segment reports by RETAC members;

• A presentation on the domestic coal market; and

• A roundtable discussion.

There is not yet an actual agenda for the upcoming meeting on the RETAC web site. There are some interesting items from the minutes of the last meeting on September 19^th, 2013:

• The Association of American Railroads reported “a performance measure of 99.998% for successful transportation of hazardous materials. Accident rates are down, resulting in rail being the safest mode of transportation. Crude oil shipment performance has been excellent, resulting in product loss of only 2.2 gallons per million ton miles.” Page 2 of Minutes. I suspect that the figures have probably changed since then.

• The Rail Fleet Update continues to show a large backlog of new bulk tank cars and further projected growth in the petroleum transportation process.

• The Rail Crude Oil Segment Report (NOTE: there is some sort of problem with data visualization on page 2; it just won’t load) contains some interesting graphics on the increase in the barrels of crude oil shipments originating in North Dakota (slide 4) and showing that 68% of the crude shipped from that region goes by rail.

The public is invited to attend the meeting, but there does not appear to be any provisions for public oral comments at the meeting. The STB is soliciting written comments; they should be emailed to Michael Higgins (Michael.Higgins@stb.dot.gov).

FRA Announces RSAC Meeting – 3-6-14

Today the DOT’s Federal Railroad Administration published a meeting announcement in the Federal Register (79 FR 9585) concerning a meeting of the Railroad Safety Advisory Committee (RSAC) on March 6^th, 2014. The RSAC meeting topics will include opening remarks from the FRA Administrator, and status reports will be provided by the Train Crew Size, Securement, Hazardous Material Issues, Fatigue Management, and Risk Reduction Working Groups.

At the August 29^th, 2013 meeting of the RSAC three Working Groups were formed to look at issues related to the hazards associated with crude oil unit trains. At the time there had been only one catastrophic rail accident associated with those trains, the Lac-Megantic, Quebec derailment in July of 2013. Since then a number of similar accidents have raised the profile of this issue. The three new Working Groups are:

• The Hazardous Materials WG;

• The Securement WG; and

• The Crew Size WG

Only the first of these three Working Groups is addressing issues that were involved in the subsequent accidents. That WG was tasked [.PDF Download Link] with reporting on the following specific issues:

• If appropriate, develop recommended regulatory language identifying criteria reflecting the types and quantities of hazardous materials (e.g., number of cars or shipments in a train, classes or identification of materials, identification of residue or loads) which are recommended to be subject to any special handling requirements or operational controls, including requirements that may be recommended as a result of this task, or Task Nos. 13-03 [Securement Issues] or 13-05 [Crew Size Issues].

• If appropriate, develop recommended regulatory language for any special handling requirements and/or operational controls for trains or vehicles meeting the criteria identified as requiring special handling and/or operational controls as a result of this task. This may include the appropriate use of wayside defect detectors, after arrival orders and track authorities, as well as crew qualifications related to emergency response activities.

• If appropriate, develop a recommendation to expand the applicability of part 172 subpart I, to require rail transportation route analysis and security plans for trains or vehicles meeting the criteria identified as requiring special handling and/or operational controls as a result of this task.

• If appropriate, develop recommended regulatory language for the accurate, efficient, and proper frequency of hazardous material testing and classification of flammable liquids, including crude oil and ethanol prior to filling/loading of rail tank cars.

• If appropriate, develop recommended regulatory language for requirements of marking of tank cars, training of personnel, and information required on shipping papers, in addition to those contained in 49 CFR part 172, subparts D, H, and C, for trains or vehicles meeting the criteria identified as requiring special handling and/or operational controls as a result of this task.

The Hazardous Materials WG has met three times since it was formed last August, the latest meeting was on January 27^th. No working group minutes or meeting notes are maintained on the RSAC web site so it is difficult to assess how far along the WG is with forming its recommendations. At the very least there will be status reports on the tasks listed above. It is not expected that the FRA will take any action on these tasks until there is a final report from the WG. That final report is due April 1^st, 2014.

Interestingly there is no mention of the October 31, 2013 meeting of the RSAC on the RSAC web site. At that meeting there was a presentation made about the status of the Hazardous Materials Working Group efforts [.PDF Download Link]. Interestingly the Hazardous Material WG notified the RSAC at that meeting that they did not plan on reporting on the route risk analysis issues because of the complexity of the issue and the time constraints involved.

There are no provisions for public comment (either oral or written) announced in the meeting notice. The public (in limited numbers) is invited to attend the meeting. Additional material will be available on the Federal eRulemaking Portal (www.Regulations.gov; Docket # FRA-2000-7257).

EO13650 and IST: A Look Back at Proposed Solutions

There has been an interesting undercurrent in the EO 13650 Working Group about the inevitable discussions about inherently safer technology that have come up in the public listening sessions being held by the Working Group. The environmental and worker’s rights advocates are predictably pushing EPA, OSHA and DHS to adopt some form of IST mandate as a way to reduce the hazards from accidental or deliberate releases of industrial chemicals. Just as obviously, industry groups continue their adamant stance against any such proposal.

The regulators that sit on the dais at these listening sessions, acting as the ears of the agencies, are placed in an unenviable position. Some level of IST commitment is almost certainly going to come out of this process as a proposal for regulations, but the regulatory community is not going to get any additional manpower or funding for implementing such a regulation because this is an Executive Order, not Congressional authorization.

Complicated Topic

All sides of the disagreement do acknowledge that inherently safer technology techniques do offer a way to reduce the hazards associated with dangerous chemicals. The two sides, however, vehemently disagree on how such a program should be implemented or how much of a role the government should have in such an implementation.

Unfortunately, neither side really listens to (or even trusts) the other side. As with most complex problems this is not a black/white issue that can be resolved by simple means. Over the last six+ years writing this blog, I have taken a hard look at the IST issue and have offered suggestions on how to effectively deal with many of the issues involved.

I have a great deal of self-confidence and think rather highly of my ideas, but I know that I am not the source of all knowledge. But I have tried to look at IST from the point of view of a professional in the chemical manufacturing industry who has spent many years making things work. Some of the ideas that I have written about could be used to form the basis of a workable IST regulatory scheme.

IST ISSUES Page

I have just put up a new page on this site that compiles a list of the IST posts that I have written over the last six years. They are not indexed; they are just listed chronologically. I have added a listing of the sub-headings from those posts (or at least the ones with such sub-headings) to give people some insight as to what is being discussed.

There are a couple of specific series of blog posts that I would like to call special attention to because they do an in-depth look at various aspects of a practical IST regulatory system. I will list the last post in these two series here as those posts will provide links to the earlier posts in the series. Those two series were:

What Standard to Apply for IST – Cost Estimates

IST Questions – Active Mitigation

Formatting Issues

I do have to apologize for the readability of some of those earlier posts. In transferring posts from the original site on AOL Communities to Google, much of the formatting disappeared. From time to time I do go back and re-edit the formatting of some of those earlier posts to make them more readable. That is a time consuming (and very boring process) so some of those posts can be more challenging to read than others. I will give priority to fixing these IST posts over some of the others, but it will still take time.

Physical Security – Another Perspective

There is an interesting piece on NewYork.CSBLocal.com detailing complaints from Sen, Schumer (D,NY) about the ability of power companies to veto security requirements. I’ll leave the discussion of the politics of electrical system security to the folks that deal with it on a routine basis. I would like to point out a serious flaw in Shumer’s reasoning, a flaw that has been bandied about quite a bit in discussions of this attack on Silicon Valley transformer stations; the assumption that these stations can be protected against a serious physical attack.

The Target

You see these stations scattered all around the country. Some of them are bigger and some are smaller, but they all have a couple of things in common. First off, the purpose of these stations is to transform the very-high voltage cross country transmission lines into the lower voltage used by local transmission networks. The equipment is widely spaced to ensure that there is room to work without worrying about sparks jumping from one set of equipment to another. The big boxes hold huge coils of copper wire surrounded by a cooling fluid, typically an oil of some sort.

If you put a hole in the outer jacket the oil drains out, the transformer overheats and at the very least shuts down. If a shooter is extremely lucky or knowledgeable the bullet will strike and damage the copper coil and cause an immediate shut down.

Another vulnerable area in these stations are the insulators that protect the physical structure of the towers and such from the high voltage transiting the lines. If these are damaged to the point that they allow arcing to the towers or fail to continue to support the transmission lines, you have another immediate shutdown of power transmission.

The slower the shut down the easier it is for the transmission company to re-route the power through other parts of the station or to other stations in a systems with a certain amount of redundancy. The more parts of the system that are taken off-line the more difficult it is to keep the power flowing. It is a testament to the system operator that the loss of 17 transformers at a single transmission station did not take down a large portion of California’s power grid.

Vulnerabilities

It seems as if this attack was executed by multiple shooters putting holes in transformer shells. This does not take snipers unless they are working from really long distances away from the site. Any trained infantryman (or hunter for that matter) can consistently hit a man-sized silhouette at about 300 meters over open sites. A transformer is a much larger target. Put a scope on the rifle and a relatively experienced shooter with a reasonable hunting rifle could engage those targets all day long out to a range of 1000 meters. A trained sniper could nearly double that distance depending on the weapon.

How hard would it be to find a firing position within 1000 meters of a transformer station? Not hard at all at any station that I have seen. Finding a concealed location would be a bit more difficult in most cases, but there really isn’t a need for concealment. These stations are typically unmanned and certainly don’t have an on-site security force to dissuade people from shooting at the facility.

There has been talk of installing a variety of ballistic barriers to protect the transformers at these sites (I suggested Kevlar® curtains on a LinkedIn page a couple of weeks ago) and these would certainly make it harder to take out a transformer (certainly a high-value target in the terms of cost and time-to-replace), but that just would not work to protect the high-voltage insulators. These are harder targets to hit, but bullets are cheap and just a little marksmanship training with a good scoped-hunting rifle would make them easy targets at reasonable ranges.

Even if you were able to harden those insulators and install blast shields, it would not take much more effort to cut a fence and place improvised explosive devices on the transformers. The only way to stop that would be to place a sizeable security team on site and put a real barrier plan into place. Even then, an experienced and determined assault team could get through. Just ask any Iraq or Afghanistan vet how hard it would be to penetrate these facilities. And remember, determined foes managed to execute routine attacks in the Green Zone in Bagdad, one of the most heavily guarded areas in the world.

Security Measures

Some sort of security measures are going to have to be put into place, but they will have to be minimal or the systems will become too costly to operate and the terrorist’s objectives will have been achieved without firing a shot. But it is not going to be possible to put in enough security to stop a determined adversary from shutting down multiple transformer stations. Perfect Security Is NOT Possible.

No the best way to deal with this potential problem is to make these areas non-targets. Make the systems so redundant that taking out one, two or even three of these stations will not be a catastrophic event. Actually, the attack outside of San Jose was one of the best security measures that we could have asked for; the utility was able to continue to supply electricity to its customers even after a significantly successful attack. This means, that as long as the Schumer’s of the world don’t over-react (which is what they do by training and inclination), most terrorist groups will have crossed these off of their target list as there are much easier targets to successfully engage.

HR 4022 Introduced – Security Clearances

As I noted last week Rep Lynch (D,MA) introduced HR 4022, the Security Clearance Reform Act of 2014. While this bill does not actually apply to security clearances provided to individuals in the private sector (other than contractors working for the federal government), this bill, if passed, should make the security clearance process more effective and in the long run, make it easier for private sector security clearances to be processed in a timely manner.

Strategic Plan

Section 2 of the bill would require the President to develop a strategic plan for updating the security clearance investigation process. Among other things that plan would address:

• The development and establishment of a continuous evaluation or monitoring system {§2(b)(1)};

• Ensuring that the background of each cleared individual is monitored on a continual basis {§2(b)(2)(A)};

• Ensuring the effective, efficient, and timely completion of background investigations relating to a covered individual’s eligibility for a security clearance {§2(b)(3)};

• Improving the procedures to require information sharing between agencies {§2(b)(4)};

• Enhancing methods for reducing or eliminating manual processes with respect to security clearance background investigations, and automating and integrating the elements of such investigations and adjudication processes {§2(b)(6)};

• Reducing or eliminating the use of databases and information sources that cannot be accessed and processed electronically, or modify such databases and information sources to enable electronic access and processing {§2(b)(7)); and

• Increasing the use of digitally processed fingerprints as a substitute for ink or paper prints to reduce error rates and improve portability of data {§2(b)(8)}.

Moving Forward

This bill is being introduced by the Ranking Member of  Federal Workforce Subcommittee of the House Oversight and Government Reform Committee and is co-sponsored by the Ranking Member of the Committee. In some committee that might ensure that the bill would get at least a hearing. I don’t think that this applies to this controversial committee.

This bill was introduced still born.

What is Toxic?

There is a heart wrenching story over at MotherJones.com about one of the many chemical response stories that came to light after the Freedom spill in Charleston, WV. This is the type story that, as a chemical professional, I cringe at every time I hear people talk about the dangers of chemicals in the environment, because this is the type story that gets repeated every time that anti-chemical activists try to put unrealistic restrictions on the chemical industry.

The Problem

My heart goes out to this poor girl and her family, but no level of research by the makers of Crude MCHM would have identified this potential problem with their product. No amount of work by the water company would have removed enough of the Crude MCHM from the water to have made it safe for this girl to bathe in or drink. There are just too many variables in the human genome to make anything safe enough for everybody.

No body suggests that we eliminate peanuts from the world, but there is a significant portion of the population that peanut allergies are a potentially fatal reality that must be dealt with every day. I had to be fed soy milk as an infant because of an apparently temporary but potentially fatal allergy to diary milk, but there was no outcry to remove cows from the planet.

Thousands of industrial chemicals are made in this country every day. New blends of those chemicals are made on almost a daily basis in response to some new industrial problem. And for the vast majority of those chemicals there is little or no real toxicological testing done. Nobody is assuming that these chemicals are safe; we in the industry tell our employees and customers not to drink these chemicals, not to allow them to get on one’s skin, not to breathe the fumes. We know that these things are not inherently ‘safe’. If you take the appropriate precautions, well documented in the safety literature, then you should not suffer any consequences.

Water Treatment

Chemicals that are used to make our drinking water safe include some of the most dangerous industrial chemicals in use today, including the periodic bogeyman chlorine gas. The hazards associated with these chemicals are well understood. They are closely monitored and tightly regulated. However, because of epidemiological reviews we are finding new hazards associated with these chemicals that could never have been identified by even the most rigorous toxicological testing. As these new hazards are identified we modify treatment regimes, we change testing standards, and we make the process and product safer.

We do not attempt to remove all ‘chemicals’ from water; first off it is not physically possible since water is itself a chemical. Even when we have to have ‘pure’ water (something that is very expensive to produce) it still contains dissolved gasses that are technically contaminants. No, drinking water contains a wide variety and concentrations of chemicals that have been determined to be safe. And it almost certainly contains an even wider variety of chemicals in extremely low concentrations that we are not aware of.

What level of water treatment is needed is determined, in large part, by the level of contaminants in the source water and what the water is used for. Again, this is as much a political decision as it is a technical decision. There is a cost for any type of water treatment and that cost will be passed along to the consumers of the water. The politicians must determine what their constituents will pay for as much as what can be treated. No politician will long survive making a decision for more expensive treatment options than are reasonably needed to clean the water going into the distribution system.

Testing Water

Testing water for contaminants is very complicated science and politics. The science of analytical chemistry is advancing by leaps and bounds (as is most of science if you really care to know). New technologies are rolled out every year making it easier to detect chemicals at levels lower than anyone ever thought of testing at before. But, this new knowledge comes at a high cost.

Some of the advances come from applying old techniques in new ways or in new combinations. Much of this is relatively easy to apply if one keeps up with all of the new information being published. Unfortunately, the people doing the testing at water treatment facilities are not analytical chemists, professionals that routinely read analytical journals and attend conferences where they share information about new techniques and equipment. No, they are professional technicians, people that take great pains at conducting repetitive tests that are designed by others.

Which tests they conduct are determined by politicians, people that make decisions on technical matters while applying cost-benefit analysis to determine which chemicals must be identified and at what level of detection they must be reported.

No lab is capable of testing for everything. Most water treatment facilities only test for a very limited number of chemicals at a relatively high level of detection; these chemicals and limits are set by law. They are required to periodically send-off samples to better equipped labs to detect a wider range of chemicals at lower levels of concentration; again these are established by law.

Adding a new chemical to the list of chemicals to be tested for is as much a political decision as it is a technical decision. First off, since testing has a cost associated with it (and that cost is variable depending on what testing technique is used), a determination must be made that there is a reasonable probability of the chemical being detected. That determination is a combination of the possibility of the chemical being in the source water for the facility, the detection limit of the tests being considered, and the amount of that chemical that might pose a safety risk in the drinking water.

Once a test is identified for a chemical that might be in the source water, a cost-benefit analysis must be done. If new, expensive equipment must be added to the lab; equipment that will require additional training and maintenance support and the chance of finding the chemical is low or the acceptable contaminant level is very high then the political decision will almost certainly be made not to implement the test in-house. For example, it appears that the water treatment facility in Charleston, WV is still relying on an outside lab to test for Crude MCHM.

What to Test For

Current federal regulations for drinking water treatment facilities require them to do a source water assessment to determine what chemicals to test for. Once a treatment authority understands what chemicals are routinely found in their source water, they determine what type of water treatment is required to ensure that those chemicals reach the distribution side of their system at safe levels. They also establish a testing regime that ensures that their facility continues to operate in a manner that meets those standards. Some of that testing will be done on site, and some will be done off site with a testing frequency that meets established standards for that type of treatment facility.

Currently, those regulations do not require a facility to take into account industrial chemicals that could possibly get into the source water in the event of a spill at a facility adjacent to the source water. This was a political decision (almost certainly an unconscious decision, or at least one that was not based upon public discussions) based upon the assumption that these facilities would have measures in place to prevent these industrial chemicals making their way into the source water.

It is now readily apparent that the source water assessment should include industrial chemicals that could reasonably (again a political term) be expected to make their way into source water for a treatment facility. That does not mean, however, that a treatment facility must be able to remove every industrial chemical that might make its way into the source water. Again, a political cost-benefit analysis must be conducted, looking at the cost of a system that could remove a chemical that has a low probability of making its way into the source water.

If the decision is made to design a facility that cannot remove that potential contaminant, then steps must be taken to ensure that the chemical does not make its way into the treatment facility. What those steps would be will depend on the chemical involved, the amount that might make its way into the source water, and how far upstream that potential chemical source is from the intake of the drinking water treatment facility.

A key component of that prevention scheme must include actions taken by the owner of the facility that houses the chemical in question. Toxic chemicals that cannot be removed by a downstream treatment facility will require higher standards of spill protection and spill notification than other chemicals at the facility. A regulatory scheme must be in place to ensure that those standards are employed and maintained.

What is Toxic?

Again, this gets back to the question of what is toxic. Anyone with any experience in the chemical industry knows that all chemicals (including water) are toxic at some level. We also know that toxicity varies substantially in the population. What is toxic for the little girl in the Mother Jones article is substantially different than that for the average healthy adult male. Establishing levels of toxicity is as much a political decision as it is a technical decision.

Again, it comes down to cost. Relatively simple toxicity testing with rats (for example) is still costly. Most chemicals that the public is not expected to come into contact with does not even require that level of testing. We can look at the chemical structure and compare that structure to other chemicals with better known toxicity profiles and determine the probability that the chemical is toxic. As long as those comparisons are done properly and the decisions are made with liberal bias towards safety, this is more than adequate for determining how those chemicals are handled in an industrial environment.

Chemicals that might reasonably be expected to contaminate source water for a drinking water treatment plant require a higher standard. For chemicals that are readily removed by a water treatment system, something that is typically easier to determine than toxicity, there should be a minimal level of toxicity testing completed to establish safe limits. More extensive testing should be required for those chemicals that the treatment facility cannot remove from the water.

Finally, epidemiological testing must be done to refine the toxicology testing. We can only truly determine what is an acceptable level of toxicity by exposing a large population to a chemical. No one will, or should, authorize such large scale testing in advance of a spill. But, once there is such a population exposure, it is only prudent to study that population to make a more accurate determination of what is actually toxic both for the chemical of exposure and related chemicals.