Water supply security concerns have been discussed at length in the United States in recent years, but in the past month this debate has taken on a new sense of urgency. The tragic events of September 11th highlighted America’s vulnerability to terrorism and spurred an unprecedented domestic security response. Water treatment facilities were identified almost immediately as a potential target for further attacks and were urged by the FBI to implement security measures, most of which are still in place.

In the short term, the security measures primarily include increased security patrols and more diligent monitoring for pathogens and chemical toxins. Long term security will necessitate investment in water storage, distribution, treatment and monitoring systems and this will create opportunities for suppliers of the respective equipment. Unfortunately however, these measures represent a rather feeble defense against any determined terrorist.

The attacks on the Pentagon and the World Trade Center were successful despite those facilities being subject to tight security. Metal detectors and armed guards were obviously no match for a suicidal pilot with a jet plane, but this was not the type of terrorism they were intended to prevent. Similarly, the implementation of security measures such as air patrols of aqueducts and ground patrols at reservoirs is unlikely to prevent a carefully planned terrorist attack. This is due partly to the scale of the task involved in patrolling a large water system and partly to the proven ineffectiveness of conventional security arrangements against an unconventional threat.

Closely monitoring water in order to catch deliberate contamination of the supply before the water reaches consumers is also more difficult than it sounds. No real-time tests exist to detect the agents most likely to be utilized by terrorists and the expense involved in frequently applying the laboratory tests that are commercially available is prohibitive. Turnaround times for laboratory tests can also be slow enough that consumption of contaminated water may occur before test results are known. While they may provide reassurance to a worried public, measures such as hiring security guards and testing samples for detectable toxins are likely to contribute little towards preventing a successful terrorist attack.

Many large water systems in the U.S. were built more than 50 years ago, when the threat of domestic bioterrorism seemed extremely remote. These systems were not built with security in mind, yet their size may be their best defense against deliberate contamination by a terrorist.

San Francisco, for example, receives part of its drinking water supply from the Crystal Springs Reservoir south of the city. Assuming perfect dilution, poisoning this reservoir with sufficient hydrogen cyanide to cause death or debilitation to someone consuming one glass of water would require over 400,000 metric tonnes of the toxin (according to a U.S. Air Force estimate of the required concentration).

Many smaller reservoirs exist, to be sure, and some contaminants can be dangerous in lower concentrations. But even the most committed terrorist would struggle to obtain and administer the volume of toxin required to contaminate an entire reservoir or aqueduct.

For this reason, attacks are more likely to focus on very small systems or small parts of larger systems. While it may be impossible to poison the entire water supply of San Francisco, it may be conceivable to poison one tank of treated San Francisco water or the entire system serving a remote desert truck stop or isolated ski resort. Not only are the volumes of toxin required more manageable for a terrorist, but security protection may be weak or nonexistent. In this case, the safety of consumers relies almost entirely on humble old chlorine.

In most cases, finished water in a tank will not be filtered again before it reaches the consumer, and even if it is this will be ineffective in many cases. Residual chlorine in the tank is the most significant remaining defense against contaminants.

While chlorine may be ineffective against chemical toxins, there are some biological toxins and pathogens against which it is effective. Cholera and the botulinum toxin that causes botulism can be inactivated by chlorine exposure. Other biological agents, however, are resistant to chlorine and the resistance of others is unknown.

Anthrax spores, for example, are resistant to chlorine and can remain stable in water for two years. The resistance of other possible bioterrorism agents such as plague and brucellosis is unknown. Even biological agents that are not resistant to chlorine can be a threat if terrorists can infiltrate a treatment plant and deactivate the chlorination system.

While tanks could be vulnerable, pipes downstream from the treatment plant also are vulnerable. Fortunately the high pressure contained in pipes makes injecting a contaminant difficult, but the relative lack of security could make this an easier target than a finished water tank. Monitoring every accessible pipe continuously is an impossibility for most water utilities so again we rely primarily on residual chlorine. Ensuring that manholes are secure and that tanks are covered and locked may be the most significant measure a water system can take to prevent this danger in the short term.

In the long term water supply safety cannot be guaranteed, but at least there are a greater number of options. In the wake of the September 11th attacks, The Association of Metropolitan Water Agencies recommended that the EPA budget for security planning be raised from $2.5 million to $155 million next year. They are also seeking an additional $5 billion to upgrade water and wastewater facilities. Some of this money would be used to improve the EPA’s notification system, which allows information to be quickly shared by water utilities, law enforcement and emergency response agencies. Some would be used to establish bioterrorism response plans and to identify current security weaknesses.

The potential for manufacturers of water treatment and monitoring equipment to benefit from increased federal funding depends to a certain degree on which areas are identified as security priorities.

Detection of chemical and biological agents is likely to be one area of emphasis. Technology already exists to enable detection of some potential threats, but until now there has been little demand for it in water monitoring applications. Examples include a gas and water quality monitor developed by Sandia National Laboratories (operated for the U.S. Department of Energy) that is capable of real-time detection of some toxic chemicals. In addition, biotechnology companies such as Cepheid, Inc. of Sunnyvale, California, have developed fast on-site tests utilizing DNA analysis for pathogens such as anthrax.

Another potential area for improvement is in the treatment facility itself, where increased use of ozonation could complement chlorination to help prevent pathogens from reaching consumers.

Ideally however, the first priority will be simply to improve security by installing tamper-proof manholes, ensuring water tanks are covered and locked, and making sure chlorination systems cannot be interfered with. This may be a relatively simple process for the nation’s 54,000 community water systems, provided they have adequate funding.

A danger exists, however, that the smaller — and therefore the most vulnerable — non-community water systems will not receive funding or sufficient information. Well over 100,000 of these systems exist in the U.S. and they serve millions of people between them. Smaller systems are the most susceptible to bioterrorism, yet they will probably receive the least assistance from the federal government.

Perhaps federal funding should be focused towards these potential bioterrorism targets rather than expensive aerial patrols over aqueducts and armed guards at major urban facilities.

This may provide fewer commercial opportunities for water treatment and monitoring equipment makers, but it will also provide fewer opportunities for terrorists.