EPA’s Treatment of Data, Models Is Worrisome

Published September 1, 2007

In late June 2007, the Environmental Protection Agency (EPA) responded to a three-year-old data quality correction request filed by the U.S. Chamber of Commerce–by far the longest time ever taken by any agency to respond to any stakeholder Information Quality petition.

The petition sought correction of fundamental physical and chemical data used in models and databases disseminated or sponsored by EPA.

Although the agency acknowledged some improvements were needed, it continues to refrain from forming an interagency committee involving other government agencies, such as the National Institute of Standards and Technology (NIST) and U.S. Geological Survey (USGS), that working together could greatly advance progress in addressing data quality problems.

Institutional Complacency

Despite the longstanding, well-known expertise of NIST and USGS in developing quality data–which was explicitly called to EPA’s attention by the Chamber and is recognized throughout the world–the agency failed to contact NIST and USGS directly, and instead contacted the co-chair of the National Science and Technology Council Digital Data interagency working group, which has no interest in specifically addressing the concerns raised in the Chamber’s petition.

EPA continues to absolve itself of addressing data and model deficiencies by issuing disclaimers with disseminated data and models.

This stance by the agency adds strength to growing stakeholder skepticism and dismay about EPA’s openness to finding ways to improve its data, assumptions, and models. Additional data quality challenges by stakeholders may be needed to correct this problem.

This matter is likely to receive intense scrutiny during the public comment period concerning EPA’s proposed ozone rulemaking.

William L. Kovacs