11 News | November 10, 2010
By Mark Greenblatt

HOUSTON—For more than 20 years, the Texas Commission on Environmental Quality under-reported the amount of radiation found in drinking water provided by communities all across Texas. As a result, health risks to people consuming the water have been underestimated in many water systems where radioactive contaminants are present.

The TCEQ regulates water systems for compliance with federal safe-water drinking regulations. However, KHOU has learned the state regulating agency consistently took radiation readings it received from the water testing lab run by the Department of State Health Services and lowered the "official" radiation readings reported by the independent lab. The TCEQ would do this by subtracting off the margin of error for all radiation readings it would receive. The subtractions helped some utilities avoid radiation violations that could have forced them to clean up their water decades ago.

Harris County Municipal Utility District No.105 is one of those utilities that benefited from the TCEQ "math." The utility did receive two official violation notices, in 2008 and 2009, for having too much radiation in the water supply it provides to thousands of residents.? However, KHOU has learned the MUD would have exceeded federal regulations for radiation in its water as far back as 1988, had the state not subtracted off the margin of error for radiation readings.?

The TCEQ confirmed in an e-mail to KHOU that MUD 105’s actual lab result in 1988 for radioactive alpha radiation came in at 17.6 picocuries (a scientific unit of measurement for radiation). The measured level came in above the federal legal limit of 15 picocuries/Liter (pCi/L) for alpha radiation in the water and would have triggered a violation.?

However, the lab’s 17.6 measurement also came with a margin of error of 5.3. That meant the lab felt the real radiation measurement for alpha radiation in 1988 could have been as high as 22.9 (5.3 points above the measured 17.6), or as low as 12.3 (5.3 points below the measured 17.6). TCEQ confirmed in an e-mail that it chose the lowest possible radiation number, 12.3, for regulatory purposes. The choice to subtract out the margin of error, instead of simply reporting the result, helped MUD 105 avoid a violation in 1988.???

TCEQ’s Linda Brookins, who oversees all drinking water safety regulation for the state of Texas, confirms the agency consistently subtracted off margin of error for water systems across Texas, since the beginning of state testing for radioactive materials in drinking water. The state began that testing more than 20 years ago. She says the TCEQ stopped the practice in 2009, after an EPA audit instructed the agency to stop subtracting margin of error from radiation readings. Brookins believes the agency’s actions did not impact human health.

A KHOU analysis of "Texas math" concludes that TCEQ’s under-reporting helped MUD 105 repeatedly avoid testing above the federal legal limits for alpha radiation in drinking water. It didn’t receive a violation until 2008, when it was found to have too much alpha radiation in its water, even after TCEQ subtracted out the margin of error.?

However, if you take away the state’s subtractions for all of its historical tests for alpha radiation, the MUD would have tested above the legal limit for alpha radiation in water at least 12 times dating back to 1988. In addition, the MUD has never received a formal violation from TCEQ for radiation in its water that comes from radium. However, nine out of the last 15 tests for radium in its water would have scored above the federal legal limit for radium (5 pCi/L) without TCEQ’s subtraction of margin of error.?

"I think, from a public health standpoint, it’s hard to defend," Dr. Joshua Hamilton said. "It’s certainly not defensible from a scientific standpoint."

Hamilton is a toxicologist and public-health scientist currently working as the chief academic and scientific officer at the Marine Biological Laboratory in Massachusetts. Hamilton received his doctorate from Cornell University in New York, has previously taught as a professor at Dartmouth Medical College and was the director of the Center for Environmental Health Sciences at Dartmouth.

"You’re not really getting an accurate picture of what your health risk might be. Nor does your community know what the actual exposures are," he said. "I don’t see how it could be accidental."

While TCEQ says it began the radiation subtractions in the 1980s, a federal rule regulating radiation in drinking water, written 10 years ago in 2000, should have put a stop to "Texas math" then. However, KHOU has learned the agency continued subtracting for nine more years until an EPA audit told them to stop.

"The word that comes to my mind is cover-up," said Dr. David Ozonoff, an environmental professor and the Chair Emeritus of Boston University’s School of Public Health. "It sure looks that way."

? Ozonoff says an easy way to understand what TCEQ did was to think of a political poll during election season. He suggests, if political pollsters measured the president’s popularity at 50 percent, plus or minus 5 percent, the president’s popularity rating would be reported as 50 percent. He says you would not report the president’s popularity as 45- or 55 percent, or risk being seen as being biased toward one political party.?

However, when it comes to radiation in drinking water, Ozonoff says, if there should have been any bias at all, it should have leaned conservatively toward protecting human health (which would have meant adding in the margin of error, if any calculations were to be performed at all).

KHOU asked TCEQ’s Brookins about it all.

KHOU: "What would you tell me if I told you that I have talked with numerous scientists across the nation that would say that what TCEQ did was bad science?"

Brookins: "Well, I guess I would have no comment on that."

"I do not believe that what TCEQ was doing at that time has impacted human health," she added.

KHOU also asked Brookins about the state’s continued subtractions for margin of error, even after the EPA published a federal rule banning the practice.

KHOU: "Did you happen to skip over page 76,727 of the federal rule? Because right here in 2000 EPA told you, ‘don’t subtract margin of error.’ Did you skip that part?"

Brookins: "It doesn’t say not to subtract."

KHOU: "It doesn’t?"

Brookins: "It is silent."

KHOU: "I’d like you to hold this in your hand for a moment and read the part underlined in blue."

Brookins: "I’m not going to do that on camera."

For the record, here is the complete text of the relevant portion we quoted from on page 76727 of the EPA’s federal rule that regulates radiation in drinking water, which was published on December 7, 2000:

5. Interpretation of Analytical Results

The Agency recognizes that States have interpreted radionuclide analytical results in a variety of ways, including adding or subtracting standard deviations from the analytical results. The Agency believes that compliance and reduced monitoring frequencies should be calculated based on the ‘‘analytical result(s)’’ as stated in § 141.26(c)(3). It is EPA’s interpretation that the analytical result is the number that the laboratory reports, not including (i.e. not adding or subtracting) the standard deviation. For example, if a laboratory reports that the gross alpha measurement for a sampling point is 7 +/- 2 pCi/L, then compliance and reduced monitoring would be calculated using a value of 7 pCi/L.