This page contains a Flash digital edition of a book.
Statistics Counts at the
Environmental Protection Agency
Barry D. Nussbaum, Chief Statistician, USEPA
entioning the U.S.
Environmental Protection
Agency might not conjure
One of the hallmarks of EPA statistics
up the image of a brigade of statisti-
is that the statistician has to get out
cians toiling behind regulations, poli-
into the real world to see how the
cies, research, daily operations, deci-
sionmaking, and enforcement actions.
data are accumulated or the surveys
However, this article will demonstrate
the many varied and crucial roles for
statisticians supporting those actions.
The examples that follow try to dem-
onstrate the wide degree of applica-
Some of the other statistical analyses seek to trace the movement
tions, methods, tools, and activities of
of chemicals through the environment. Mathematical mixing mod-
the EPA statistician. By the way, there
els use information from stable isotopes to quantify the propor-
are a lot of examples that follow, but
tion of different sources contributing to air or water bodies, food
they are by no means exhaustive. There is always more to do!
sources to animal diets, and soil horizons to plant water uptake.
In the water pollution area, some statistical applications are the
Teamwork Counts derivation of effluent limits for industrial facilities, the assessment
Before giving a sampling of the EPA’s statistical activities, it is
of fluid runoff from airline deicing operations, the quantification
imperative to mention that the statistician does not work in a
of the effect of unused pharmaceuticals, the calculation of fish and
vacuum. Far from it, the statistician (or at least the successful one)
water consumption estimates, and the estimation of the concentra-
is part of a team that encompasses many areas of expertise, such
tion of pollutants in biosolids. As my colleagues inform me—this
as biology, engineering, chemistry, economics, law, ecology, toxi-
last application to view the biosolids, which is output after munici-
cology, and demography. All this is necessary as each analyst must
pal wastewater is treated—may not be the coolest of field trips ever,
understand the nomenclature, nuances, and concerns of the other
but a necessary evil.
disciplines. To support the EPA, the team must be knowledgeable in
In the air arena, EPA statisticians are busy measuring, analyzing,
the regulatory, policy, economic, legal, technical, and enforcement
and reporting the trends in pollutants from both local and nation-
aspects of the problem. So, EPA’s statisticians do make the conscien-
al perspectives, with their associated impact on the attainment of
tious effort to contribute more than just their own specialty.
health-based air pollution standards. They assess the degree of com-
pliance with auto emission standards and the efficacy of emission
The Programs
inspection programs. One interesting cross media application is the
A great deal of the EPA’s scientific research including statistics is current effort to demonstrate how large air quality computer mod-
devoted to determining the extent of the response to environmental els built for regulatory applications can be used to improve expo-
hazards by analyzing dose-response relationships. This is particu- sure estimates in epidemiologic studies. And, yes, we are involved
larly tricky as one tries to calculate responses to lower and lower with fuel economy estimates for new vehicles, a topic that endlessly
doses of toxics, some of which even fall below detection capabili- tests our ability to explain harmonic means.
ties of modern instrumentation. Statisticians try to estimate these Pesticides programs include performing the human health risk
responses, including the evaluation of uncertainty. They fit linear assessments—including characterizing and assessing exposure and
and nonlinear models to animal dose-response data to get a “best risks to humans through the food, drinking water, and residential
fit.” Even more taxing is the attempt to calculate cumulative risk use exposure pathways. Some of the challenges here involve recog-
from hazardous toxics. To an increasing extent, Bayesian inference is nizing that environmental exposures are low dose and ill defined.
being used. Sometimes the problems are confined to certain human Therefore, a series of multilevel, multidisciplinary studies, as well
subpopulations. For instance, two prominent studies concern chil- as Monte Carlo analyses, are used to establish the link between
dren’s health. One tries to derive the relationship between childhood exposure and disease.
leukemia and environmental pollutants. The other seeks to identify
Evaluation of toxic chemicals is a major EPA effort. Here, the
the causes of the continued prevalence of toxic lead poisoning in
EPA statisticians must design adequate sampling plans for soil with
children, despite the elimination of lead from paint and gasoline.
wide fluctuations in the data values. In another cross-discipline
AMSTAT November 08.indd 23 10/24/08 2:28:17 PM
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80  |  Page 81  |  Page 82  |  Page 83  |  Page 84
Produced with Yudu -