&EPA
United States
Environmental Protection
Agency
PA's Report on the Environment
Indicators Presenting Data for
EPA Region 10
-------
-------
ROE Indicators Presenting Data for EPA Region 10
EPA's 2008 Report on the Environment (ROE) compiles, in one place,
the most reliable indicators currently available to answer 23 questions
that EPA believes are of critical importance to its mission to protect
human health and the environment.
The indicators are supported by data gathered from federal and state
agencies and non-governmental organizations. All ROE indicators were
peer-reviewed to meet exacting standards for accuracy,
representativeness, and reliability, and the report has undergone
extensive internal and interagency review, SAB review, and public
comment. The indicators present trends wherever adequate data are
available, and establish national baselines where they are not. The report
also identifies key limitations of these indicators and gaps where reliable
indicators do not yet exist. The report does not propose actions to reduce
data limitations, fill gaps, or analyze the costs and benefits of doing so.
The ROE presents 78 environmental indicators at the national scale. In
response to recommendations from within and outside EPA, 32 of these
indicators also present information at regional scales, most of them by
EPA Region. There are also seven special pilot indicators developed by
EPA Regional Offices that cover only one EPA Region or parts of one
or two EPA Regions.
The table on the next page lists (1) all of the ROE indicators that present
data relevant to EPA Region 10, and (2) the ROE questions they help to
answer. Copies of all of the indicators listed are provided following the
table. Additional information on the ROE is available at
www.epa.gov/roe.
-------
Indicators with Data Specific to EPA Region 10
Air Chapter
Outdoor Air Quality
What are the trends in outdoor air quality and their effects on human health and the
environment?
Carbon Monoxide Emissions
Ambient Concentrations of Carbon Monoxide
Nitrogen Oxides Emissions
Ambient Concentrations of Nitrogen Dioxide
Volatile Organic Compounds Emissions
Ambient Concentrations of Ozone
Ozone Injury to Forest Plants
Particulate Matter Emissions
Ambient Concentrations of Particulate Matter
Regional Haze
Sulfur Dioxide Emissions
Acid Deposition
Percent of Days with Air Quality Index Values Greater Than 100
Air Toxics Emissions
Greenhouse Gases
What are the trends in greenhouse gas emissions and concentrations?
No indicators with data specific to EPA Region 10
Indoor Air Quality
What are the trends in indoor air quality and their effects on human health?
No indicators with data specific to EPA Region 10
Page
7
9
11
13
15
17
20
22
26
30
32
34
39
42
-------
Indicators with Data Specific to EPA Region 10 (cont.)
Water Chapter
Water and Watersheds
What are the trends in the extent and condition of fresh surface waters and their effects on
human health and the environment?
High and Low Stream Flows
Nitrogen and Phosphorus Loads in Large Rivers
Benthic Macroinvertebrates in Wadeable Streams
Ground Water
What are the trends in the extent and condition of ground water and their effects on human
health and the environment?
No indicators with data specific to EPA Region 10
Wetlands
What are the trends in the extent and condition of wetlands and their effects on human health
and the environment?
No indicators with data specific to EPA Region 10
Coastal Waters
What are the trends in the extent and condition of coastal waters and their effects on human
health and the environment?
Trophic State of Coastal Waters
Coastal Sediment Quality
Coastal Benthic Communities
Coastal Fish Tissue Contaminants
Drinking Water
What are the trends in the quality of drinking water and their effects on human health?
Population Served by Community Water Systems with No Reported Violations of Health-Based
Standards
Recreational Waters
What are the trends in the condition of recreational waters and their effects on human health
and the environment?
No indicators with data specific to EPA Region 10
Page
45
49
52
55
59
62
64
67
-------
Indicators with Data Specific to EPA Region 10 (cont.)
Consumable Fish and Shellfish
What are the trends in the condition of consumable fish and shellfish and their effects on human
health?
Coastal Fish Tissue Contaminants
Land Chapter
Land Cover
What are the trends in land cover and their effects on human health and the environment?
Land Cover
Forest Extent and Type
Land Cover in the Puget Sound/Georgia Basin
Land Use
What are the trends in land use and their effects on human health and the environment?
Land Use
Urbanization and Population Change
Wastes
What are the trends in wastes and their effects on human health and the environment?
No indicators with data specific to EPA Region 10
Chemicals Used on the Land
What are the trends in chemicals used on the land and their effects on human health and the
environment?
Fertilizer Applied for Agricultural Purposes
Contaminated Land
What are the trends in contaminated land and their effects on human health and the
environment?
No indicators with data specific to EPA Region 10
64
70
94
74
76
81
84
-------
Indicators with Data Specific to EPA Region 10 (cont.)
Human Exposure and Health Chapter
Exposure to Environmental Contaminants
What are the trends in human exposure to environmental contaminants, including across
population subgroups and geographic regions?
No indicators with data specific to EPA Region 10
Health Status
What are the trends in health status in the United States?
No indicators with data specific to EPA Region 10
Disease and Conditions
What are the trends in human disease and conditions for which environmental contaminants
may be a risk factor, including across population subgroups and geographic regions?
Cardiovascular Disease Prevalence and Mortality
Chronic Obstructive Pulmonary Disease Prevalence and Mortality
Ecological Condition Chapter
Extent and Distribution
What are the trends in the extent and distribution of the nation's ecological systems?
Land Cover
Forest Extent and Type
Forest Fragmentation
Land Use
Urbanization and Population Change
Land Cover in the Puget Sound/Georgia Basin
Diversity and Biological Balance
What are the trends in the diversity and biological balance of the nation's ecological systems?
Coastal Benthic Communities
Benthic Macroinvertebrates in Wadeable Streams
Fish Fauna! Intactness
Non-Indigenous Benthic Species in the Estuaries of the Pacific Northwest
Page
87
91
70
94
97
76
81
74
62
52
99
102
-------
Indicators with Data Specific to EPA Region 10 (cont.)
Ecological Processes
What are the trends in the ecological processes that sustain the nation's ecological systems?
Carbon Storage in Forests
Physical and Chemical Attributes
What are the trends in the critical physical and chemical attributes of the nation's ecological
systems?
U.S. and Global Mean Temperature and Precipitation
High and Low Stream Flows
Sea Level
Nitrogen and Phosphorus Loads in Large Rivers
Ecological Exposure to Contaminants
What are the trends in biomarkers of exposure to common environmental contaminants in
plants and animals?
Coastal Fish Tissue Contaminants
Ozone Injury to Forest Plants
Page
104
106
45
110
49
64
20
-------
INDICATOR
Carbon Monoxide Emissions
Carbon monoxide (CO) gas forms primarily when
carbon fuels are not burned completely. Mobile
sources account for the majority of CO emissions (U.S.
EPA, 2003). These sources include both on-road vehicles
(e.g., cars, trucks, motorcycles) and nonroad vehicles and
engines (e.g., farm equipment, construction equipment,
aircraft, marine vessels). Consequently, high concentra-
tions of CO generally occur in areas with heavy traffic
congestion. In cities, as much as 95 percent of all CO
emissions may come from automobile exhaust (U.S. EPA,
2003). Other sources of CO emissions include industrial
processes, non-transportation fuel combustion, and natural
sources, such as wildfires. Fuel-burning appliances also
are a large source of CO releases in indoor environments.
Undetected releases of carbon monoxide in indoor settings
can present serious health risks to building occupants. The
CO Concentrations indicator describes health hazards
associated with inhaling CO.
This indicator presents CO emissions from tradition-
ally inventoried anthropogenic source categories: (1) "Fuel
combustion," which includes emissions from coal-, gas-,
and oil-fired power plants and industrial, commercial, and
institutional sources, as well as residential heaters (e.g.,
•wood-burning stoves) and boilers; (2) "Other industrial
processes," which includes chemical production, petro-
leum refining, metals production, and industrial processes
other than fuel combustion; (3) "On-road vehicles,"
•which includes cars, trucks, buses, and motorcycles; and
(4) "Nonroad vehicles and engines," such as farm and
construction equipment, lawnmowers, chainsaws, boats,
ships, snowmobiles, aircraft, and others. The indicator
also includes estimates of biogenic CO emissions in 2002.
Biogenic emissions were estimated using the Biogenic
Emissions Inventory System Model, Version 3.12, with
data from the Biogenic Emissions Landcover Database and
2001 annual meteorological data.
CO emissions data are tracked by the National Emis-
sions Inventory (NEI). The NEI is a composite of data from
many different sources, including industry and numerous
state, tribal, and local agencies. Different data sources use
different data collection methods, and many of the emissions
data are based on estimates rather than actual measurements.
For most fuel combustion sources and industrial sources,
emissions are estimated using emission factors. Emissions
from on-road and nonroad sources were estimated using
EPA-approved modeling approaches (U.S. EPA, 2007a).
NEI data have been collected since 1990 and cover all
50 states and their counties, D.C., the U.S. territories of
Puerto Rico and Virgin Islands, and some of the territories
of federally recognized American Indian nations. Data are
presented for 1990 and from 1996 to 2002; prior to 1996,
only the 1990 data have been updated to be comparable to
the more recent inventories.
Exhibit 2-1. CO emissions in the U.S. by
source category, 1990 and 1996-2002
160
_140
^n
i 120
J 100
1. 80
o>
| 60
'o>
•| 40
m 20
0
A. Anthropogenic CO emissions by source category3
Fuel combustion
On-road vehicles
Nonroad vehicles and engines
'90
'96 '97 '98 '99 '00 '01 '02
Year
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
all fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
Data source: U.S. EPA, 2007b
B. Relative amounts of CO
emissions from anthropogenic
and biogenic sources, 2002
Biogenic
5%
Anthropogenic
95%
What the Data Show
This indicator focuses on trends in CO emissions from
anthropogenic sources. However, CO emissions from bio-
genic sources were estimated for 2002 to provide a sense of
the relative contributions of natural versus anthropogenic
emissions (Exhibit 2-1, panel B). Nationally, biogenic
emissions were estimated to contribute approximately 5
percent to the CO emissions from all sources during 2002.
Nation-wide estimated anthropogenic CO emissions
have decreased 35 percent between 1990 and 2002, the
most recent year for which aggregate NEI emissions
estimates are available (Exhibit 2-1, panel A). Almost
the entire emissions reduction is attributed to decreased
emissions from on-road mobile sources. In 2002, mobile
sources (both on-road and nonroad sources combined)
accounted for 90 percent of the nation's total anthropo-
genic CO emissions. The CO emissions reductions are
reflected in corresponding reductions in ambient concen-
trations (the CO Concentrations indicator).
Net estimated anthropogenic CO emissions declined
in all EPA Regions between 1990 and 2002 (Exhibit
2-2). The largest decrease (10.84 million tons) occurred
in Region 9, and the smallest decrease (1.33 million tons)
occurred in Region 10.
-------
INDICATOR
Carbon Monoxide Emissions
Indicator Limitations
• Comparable CO emissions estimates through the NEI are
available only for 1990 and 1996-2002. Data for 1991-1995
are not provided due to differences in emissions estimation
methodologies from other inventory years, which could
lead to improper trend assessments.
• CO emissions from "miscellaneous sources," including
\vildfires, are not included in the total emissions. Yearly
fluctuations in wildfire emissions have the potential to
mask trends in anthropogenic emissions and therefore have
been excluded from the trends graphics. Details on emis-
sions from miscellaneous sources can be found by down-
loading 2002 NEI inventory data for the "nonpoint sector"
(http://www. epa.gov/ttn/chief/net/2002inventory.html).
• The emissions data for CO are largely based on estimates
that employ emission factors generated from empirical
and engineering studies, rather than on actual measure-
ments of CO emissions. Although these estimates are
generated using -well-established approaches, the esti-
mates have uncertainties inherent in the emission factors
and emissions models used to represent sources for which
emissions have not been directly measured.
• The methodology for estimating emissions is continually
reviewed and is subject to revision. Trend data prior to
any revisions must be considered in the context of
those changes.
• Not all states and local agencies provide the same data or
level of detail for a given year.
Data Sources
Summary data in this indicator were provided by EPA's
Office of Air Quality Planning and Standards, based on
biogenic and anthropogenic CO emissions data in the
NEI (U.S. EPA, 2007b) (http://www.epa.gov/ttn/chief/
net/2002inventory.html). This indicator aggregates the
NEI data by source type (anthropogenic or biogenic),
source category, and EPA Region.
References
U.S. EPA (United States Environmental Protection
Agency). 2007a. Documentation for the final 2002 mobile
National Emissions Inventory, Version 3.
U.S. EPA. 2007b. Data from the 2002 National Emissions
Inventory, Version 3.0. Accessed 2007.
U.S. EPA. 2003. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
Exhibit 2-2. CO emissions in the U.S. by
EPA Region, 1990 and 1996-20023
30
_ 25
tf>
c
o
c 20
O
I 15
o>
c
110
E
m 5
0
-R1
—R2
-R3
R4
—R5
R6
R7
R8
—R9
-R10
'96 '97 '98 '99 '00 '01 '02
Year
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
all fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
Data source: U.S. EPA, 2007b
EPA Regions
-------
Ambient Concentrations of Carbon Monoxide
Carbon monoxide (CO) gas forms primarily when car-
bon fuels are not burned completely. Elevated ambient
air concentrations of CO are hazardous because inhaled
CO enters the bloodstream and reduces the amount of
oxygen that the blood can deliver to the body's organs and
tissues. If exposure concentrations are high enough, poten-
tially serious cardiovascular and neurological effects can
result. Visual impairment, reduced work capacity, reduced
manual dexterity, poor learning ability, and difficulty in
performing complex tasks are all associated with exposure
to elevated CO levels (U.S. EPA, 2000).
Motor vehicle exhaust currently accounts for the
majority of CO emissions nation-wide, and as much as
95 percent of CO emissions in cities with high traffic
congestion. Other anthropogenic sources of CO emis-
sions include fossil fuel combustion for heating and power
generation, metals processing, and chemical manufactur-
ing. The highest ambient air concentrations of CO often
occur during nighttime inversion conditions, which trap
pollutants near ground level. These conditions are most
frequently observed during the cold winter months (U.S.
EPA, 2003).
This indicator presents ambient CO concentrations
in parts per million (ppm) from 1980 to 2006, based on
continuous measurements averaged over 8-hour time
frames. The 8-hour standard is indicative of exposures
occurring over a sustained period of time, for example,
an outdoor -worker's exposure over the course of a -work
day. This indicator displays trends in the annual second
highest 8-hour CO concentrations for 144 sites in 102
counties nation-wide that have consistent data for the
period of record in the State and Local Air Monitoring
Stations net-work or by other special purpose monitors.
It also shows trends in the average 8-hour measurements
in each EPA Region. This indicator's exhibits display the
National Ambient Air Quality Standard (NAAQS) for
CO as a point of reference, but the fact that the national or
any regional second highest 8-hour values fall below the
standard does not mean that all monitoring sites nation-
ally or in the EPA Region also are below the standard.
The indicator displays trends in the number of the 144 sites
nation-wide at -which reported CO concentrations -were
above the level of the 8-hour standard, but this statistic is
not displayed for each EPA Region.
What the Data Show
The 2006 annual second highest 8-hour CO concentra-
tion averaged across 144 monitoring sites nation-wide -was
75 percent lower than that for 1980, and is the lowest
level recorded during the past 27 years (Exhibit 2-3,
panel A). The downward trend in CO concentrations in
the 1990s parallels the downward trend observed in CO
emissions, -which has been attributed largely to decreased
emissions from mobile sources (the CO Emissions indica-
Exhibit 2-3. Ambient CO concentrations in the
U.S., 1980-2006a
A. Ambient concentrations
90% of sites have concentrations below this line
10% of sites have
concentrations below this line
'90 '92 '94 '96
Year
'00 '02 '04 '06
B. Number of trend sites above NAAQS
90 '92 '94 '96
Year
'00 '02 '04 '06
Coverage: 144 monitoring sites in 102 counties nationwide (out of
a total of 375 sites measuring CO in 2006) that have sufficient
data to assess CO trends since 1980.
Data source: U.S. EPA, 2007
tor). In addition, of the 144 sites used to determine this
trend (out of 375 total monitoring sites that -were operat-
ing in 2006), the number reporting CO concentrations
above the level of the CO standard declined to zero over
the same period (Exhibit 2-3, panel B).
Also shown in Exhibit 2-3 (panel A) are the 90th and
10th percentiles based on the distribution of annual statistics
at the monitoring sites. This provides additional graphical
representation of the distribution of measured concentra-
tions across the monitoring sites for a given year. Thus, the
graphic displays the concentration range -where 80 percent
of measured values occurred for that year.
Consistent -with the nation-wide trend, CO levels in all
ten EPA Regions have steadily decreased since 1980, -with
percent reductions over this period ranging from 68 per-
cent (Region 7) to 85 percent (Region 1) (Exhibit 2-4).
-------
Ambient Concentrations of Carbon Monoxide
Exhibit 2-4. Ambient CO concentrations in the
contiguous U.S. by EPA Region, 1980-20063
NAAQS = 9 ppm
'82
'90 '92 '94 '96 '98 '00 '02 '04 '06
Year
Coverage: 141 monitoring sites
in the EPA Regions (out of a total
of 375 sites measuring CO in
2006) that have sufficient data to
assess CO trends since 1980.
Data source: U.S. EPA, 2007
EPA Regions
Indicator Limitations
• Because most CO monitoring sites are located in high-
traffic urban areas, the nation-wide trends presented in
this indicator might not accurately reflect conditions
outside the immediate urban monitoring areas.
• Because of the relatively small number of trend sites in
some EPA Regions, the regional trends are subject to
greater uncertainty than the national trends. Some EPA
Regions with low average concentrations may include
areas with high local concentrations, and vice versa.
• To ensure that long-term trends are based on a consistent
set of monitoring sites, selection criteria were applied to
identify the subset of CO monitoring sites with sufficient
data to assess trends since 1980. Monitoring sites -with-
out sufficient data are not included in the trend analysis.
Some excluded monitoring sites reported CO concentra-
tions above the level of the CO standard over the time
frame covered by this indicator. In 2006, for example,
one monitoring site in the U.S. recorded CO concentra-
tions above the level of the NAAQS, but did not have
sufficient long-term data to be considered a trend site for
this indicator.
Data Sources
Summary data in this indicator -were provided by EPA's
Office of Air Quality Planning and Standards, based on
CO ambient air monitoring data in EPA's Air Quality
System (U.S. EPA, 2007) (http://www.epa.gov/ttn/airs/
airsaqs/). National and regional trends in this indicator are
based on the subset of CO monitoring stations that have
sufficient data to assess trends since 1980.
References
U.S. EPA (United States Environmental Protection
Agency). 2007. Data from the Air Quality System.
Accessed 2007.
U.S. EPA. 2003. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
U.S. EPA. 2000. Air quality criteria for carbon monoxide,
2000. EPA/600/P-99/001F. Research Triangle Park, NC.
10
-------
INDICATO
Nitrogen Oxides Emissions
l\ I
I N
itrog611 oxides" (NOJ is the term used to describe the
sum of nitric oxide (NO), nitrogen dioxide (NO ), and
other oxides of nitrogen. Most airborne NO comes from
combustion-related emissions sources of human origin,
primarily fossil fuel combustion in electric utilities, high-
temperature operations at other industrial sources, and
operation of motor vehicles. However, natural sources, like
biological decay processes and lightning, also contribute to
airborne NO . Fuel-burning appliances, like home heat-
ers and gas stoves, produce substantial amounts of NO in
indoor settings (U.S. EPA, 2003).
NO plays a major role in several important environ-
mental and human health issues. Short-term and long-term
exposures to elevated air concentrations of NO are associ-
ated with various acute and chronic respiratory effects (U.S.
EPA, 1993). NO and volatile organic compounds react in
the presence of sunlight to form ozone, which also is associ-
ated with human health and ecological effects (the Ozone
Concentrations indicator). NO and other pollutants react in
the air to form compounds that contribute to acid deposi-
tion, which can damage forests and cause lakes and streams
to acidify (the Acid Deposition indicator). Deposition of
NO also affects nitrogen cycles and can contribute to nui-
sance growth of algae that can disrupt the chemical balance
of nutrients in water bodies, especially in coastal estuaries
(the Lake and Stream Acidity indicator; the Trophic State
of Coastal Waters indicator). NO also plays a role in several
other environmental issues, including formation of particu-
late matter (the PM Concentrations indicator), decreased
visibility (the Regional Haze indicator), and global climate
change (the U.S. Greenhouse Gas Emissions indicator; the
Greenhouse Gas Concentrations indicator).
This indicator presents NO emissions from tradition-
ally inventoried anthropogenic source categories: (1) "Fuel
combustion: selected power generators," which includes
emissions from coal-, gas-, and oil-fired power plants
that are required to use continuous emissions monitors
(CEMs) to report emissions as part of the Acid Rain Pro-
gram (ARP); (2) "Fuel combustion: other sources," which
includes industrial, commercial, and institutional sources,
as well as residential heaters and boilers not required to use
CEMs; (3) "Other industrial processes," which includes
chemical production and petroleum refining; (4) "On-road
vehicles," which includes cars, trucks, buses, and motor-
cycles; (5) "Nonroad vehicles and engines," such as farm and
construction equipment, lawnmowers, chainsaws, boats,
ships, snowmobiles, aircraft, and others. Since a substantial
portion of airborne NO comes from fossil fuel combus-
tion in electric utilities, this indicator includes the separate
category for "selected power generators" in addition to the
four categories presented in the other emissions indicators.
The indicator also includes estimates of biogenic NO emis-
sions in 2002. Biogenic emissions were estimated using the
Exhibit 2-7. NOX emissions in the U.S. by
source category, 1990 and 1996-2002
A. Anthropogenic NOX emissions by source category3
30
_ 30
| 25
o
= 20
"« 15
o
'« 10
'E
m 5
n
-
~
—K
Other industrial processes
Fuel combustioiTr~~~- • . X
selected power generators \
Fuel combustion: other sources \
On-road vehicles
Nonroad vehicles and engines
'90
'96 '97
'99 '00
'01
'02
Year
B. Relative amounts of NOX
emissions from anthropogenic
and biogenic sources, 2002
Biogenic
5%
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
bThis category includes
emissions from only those
power plants required to use continuous emissions monitors under the
Acid Rain Program.
Data source: U.S. EPA, 2007b
Biogenic Emissions Inventory System Model, Version 3.12,
•with data from the Biogenic Landcover Database and 2001
annual meteorological data.
NO emissions data are tracked by the National Emis-
sions Inventory (NEI). The NEI is a composite of data
from many different sources, including industry and
numerous state, tribal, and local agencies. Different data
sources use different data collection methods, and many
of the emissions data are based on estimates rather than
actual measurements. For major electricity generating
units, most data come from CEMs that measure actual
emissions. For other fuel combustion sources and indus-
trial processes, data are estimated using emission factors.
Emissions from on-road and nonroad sources were esti-
mated using EPA-approved modeling approaches (U.S.
EPA, 2007a).
NEI data have been collected since 1990 and cover all
50 states and their counties, D.C., the U.S. territories of
Puerto Rico and Virgin Islands, and some of the territories
of federally recognized American Indian nations. Data are
presented only for 1990 and the years from 1996 to 2002;
11
-------
INDICATO
Nitrogen Oxides Emissions
prior to 1996, only the 1990 data have been updated to be
comparable to the more recent inventories.
What the Data Show
This indicator focuses on trends in NO emissions from
X
anthropogenic sources. However, NO emissions from
biogenic sources were estimated for 2002 to provide a sense
of the relative contributions of natural versus anthropogenic
emissions. Nationally, biogenic emissions were estimated to
contribute approximately 5 percent to NO emissions from
all sources during 2002 (Exhibit 2-7, panel B).
According to the NEI data, estimated nation-wide
anthropogenic emissions of NO decreased by 17 percent
between 1990 and 2002 (from 25,160,000 to 20,917,000
tons) (Exhibit 2-7, panel A). This downward trend results
primarily from emissions reductions at electric utilities and
among on-road mobile sources. Although total nation-
wide anthropogenic NO emissions decreased during this
period, emissions from some sources (such as nonroad
vehicles and engines) have increased since 1990.
Estimated anthropogenic NO emissions in nine of
the ten EPA Regions decreased between 1990 and 2002
(Exhibit 2-8). The percent change in emissions over this
time frame ranged from a 36 percent decrease (in Region
2) to a 6 percent increase (in Region 10), and the largest
absolute reduction (919,000 tons) occurred in Region 3.
Indicator Limitations
• Comparable NO emissions estimates through the NEI
are available only for 1990 and 1996-2002. Data for
1991-1995 are not provided due to differences in emis-
sions estimation methodologies from other inventory
years, which could lead to improper trend assessments.
• NO emissions from miscellaneous sources are not
X
included in the total emissions.
• Though NO emissions from most electric utilities are
measured directly using continuous monitoring devices,
NO emissions data for most other source types are
estimates. These estimates are generated using -well-
established approaches, but still have uncertainties inher-
ent in the emission factors and emissions models used
to represent sources for -which emissions have not been
directly measured.
• The methodology for estimating emissions is continually
reviewed and is subject to revision. Trend data prior to
any revisions must be considered in the context of
those changes.
• Not all states and local agencies provide the same data or
level of detail for a given year.
Exhibit 2-8. NOX emissions in the U.S. by
Region, 1990 and 1996-20023
EPA
•& 2
E
LLJ
1
0
-R1
—R2
—R3
R4
—R5
R6
R7
R8
—R9
—R10
'90
'96 '97 '98 '99 '00 '01 '02
Year
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
Data source: U.S. EPA, 2007b
EPA Regions
Data Sources
Summary data in this indicator -were provided by EPA's
Office of Air Quality Planning and Standards, based on
anthropogenic and biogenic NOx emissions data in EPA's
NEI (U.S. EPA, 2007b) (http://www.epa.gov/ttn/chief/
net/2002inventory.html). This indicator aggregates the NEI
data by source type (anthropogenic or biogenic), source
category, and EPA Region.
References
U.S. EPA (United States Environmental Protection
Agency). 2007a. Documentation for the final 2002 mobile
National Emissions Inventory, Version 3.
U.S. EPA. 2007b. Data from the 2002 National Emissions
Inventory, Version 3.0. Accessed 2007.
U.S. EPA. 2003. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
U.S. EPA. 1993. Air quality criteria for oxides of nitrogen.
EPA/600/8-91/049aF-cF. Research Triangle Park, NC.
12
-------
INDICATO
Ambient Concentrations of Nitrogen Dioxide
Nitrogen dioxide (NO2) is a reddish-brown, highly reac-
tive gas that is formed in the ambient air through the
oxidation of nitric oxide (NO). Nitrogen dioxide is one
in a group of highly reactive gases generically referred to
as "nitrogen oxides" (NO ), all of which contain nitrogen
and oxygen in varying amounts. NO plays a major role in
the formation of ozone in the atmosphere through a com-
plex series of reactions with volatile organic compounds.
NO2 is the most -widespread and commonly found nitro-
gen oxide (U.S. EPA, 2003).
Short-term exposures (e.g., less than 3 hours) to low
levels of NO may lead to changes in airway responsiveness
and lung function in individuals with preexisting respira-
tory illnesses. These exposures may also increase respira-
tory illnesses in children. Long-term exposures to NO2
may lead to increased susceptibility to respiratory infection
and may cause irreversible alterations in lung structure
(U.S. EPA, 1995).
Atmospheric transformation of NO can lead to the
formation of ozone and nitrogen-bearing particles (e.g.,
nitrates, nitric acid). Deposition of nitrogen can lead to fer-
tilization, eutrophication, or acidification of terrestrial, -wet-
land, and aquatic (e.g., fresh water bodies, estuaries, coastal
•water) systems. These effects can alter competition among
existing species, leading to changes in species abundance and
distribution within communities. For example, eutrophic
conditions in aquatic systems can produce explosive growth
of algae leading to hypoxia or an increase in levels of toxins
harmful to fish and other aquatic life (U.S. EPA, 1993).
This indicator presents ambient NO concentrations in
parts per million (ppm) from 1980 to 2006, based on the
annual arithmetic average. The indicator displays trends
averaged over 87 sites in 64 counties nation-wide that have
consistent data for the period of record in the State and
Local Air Monitoring Stations net-work or by special pur-
pose monitors. It also shows trends in the annual average
NO measurements in each EPA Region. This indicator's
exhibits display the NO2 National Ambient Air Quality
Standard (NAAQS) as a point of reference, but the fact that
the national or any regional average values fall below the
standard does not mean that all monitoring sites nation-
ally or in the EPA Region also are below the standard.
This indicator displays trends in the number of the 87 sites
nation-wide at -which NO concentrations exceeded the level
of the annual average standard over the period of record, but
this statistic is not displayed for each EPA Region.
What the Data Show
The national annual average NO concentration in 2006
•was 41 percent lower than that recorded in 1980 (Exhibit
2-9, panel A). Also shown on this graph are the 90th and
10th percentiles of NO concentrations based on the distri-
bution of annual statistics at the monitoring sites. This pro-
vides additional graphical representation of the distribution
Exhibit 2-9. Ambient NC>2 concentrations in the
U.S., 1980-20063
A. Ambient concentrations
0.00
10% of sites have
concentrations below this line
'90 '92 '94 '96
Year
'02 '04 '06
B. Number of trend sites above NAAQS
<= £
92
co o Q-
£
'90 '92 '94
Year
'02 '04 '06
Coverage: 87 monitoring sites in 64 counties nationwide (out of a
total of 369 sites measuring NCb in 2006) that have sufficient data
to assess N02 trends since 1980.
Data source: U.S. EPA, 2007
of measured concentrations across the monitoring sites for
a given year. Thus, for each year, the graphic displays the
concentration range -where 80 percent of measured values
occurred. The highest annual average NO concentra-
tions are typically found in urban areas. In addition, of
the 87 sites used to determine this trend (out of 369 total
monitoring sites that -were operating in 2006), the number
reporting NO2 concentrations above the level of the NO2
standard declined from seven sites in 1981 to zero sites
since 1992 (Exhibit 2-9, panel B).
NO levels in all ten EPA Regions have steadily
decreased since 1980, -with percent reductions over this
time ranging from 20 percent in Region 8 to 49 percent in
Region 1 (Exhibit 2-10).
13
-------
INDICATO
Ambient Concentrations of Nitrogen Dioxide
The decrease in NO2 concentrations in this indicator is
consistent with the decreasing NO emissions observed over
the past decade (the Nitrogen Oxides Emissions indicator).
Indicator Limitations
• Because ambient monitoring for NO occurs almost
exclusively in high-traffic urban areas, the average
concentrations presented in this indicator likely may not
reflect NO2 levels in rural areas. Also, in rural areas, air
mass aging could foster greater relative levels of peroxy-
acetyl nitrate (PAN) and nitric acid which can cause a
positive interference in NO2 measurements.
• The measurement of NO2 is based on the conversion of
NO2 to NO and the subsequent detection of NO using
the chemiluminescence technique. Because there are
other nitrogen-containing compounds, such as PAN and
nitric acid, that can be converted to NO, the chemilu-
minescence technique may overestimate NO2 concentra-
tions due to these interferences. Measurement devices
•with ultraviolet photolytic converters are less prone to
interferences than devices with heated surfaces (or cata-
lysts) upstream of the chemiluminescence detector.
• Because of the relatively small number of trend sites in
some EPA Regions, the regional trends are subject to
greater uncertainty than the national trends. Some EPA
Regions with low average concentrations may include
areas with high local concentrations, and vice versa.
• To ensure that long-term trends are based on a con-
sistent set of monitoring sites, selection criteria were
applied to identify the subset of NO2 monitoring sites
•with sufficient data to assess trends since 1980. Monitor-
ing sites \vithout sufficient data are not included in the
trend analysis. Some excluded monitoring sites reported
NO2 concentrations above the level of the NO2 standard
over the time frame covered by this indicator. In 2006,
however, no monitoring sites in the U.S. measured NO2
concentrations above the level of the NAAQS.
Data Sources
Summary data in this indicator were provided by EPA's
Office of Air Quality Planning and Standards, based on
NO ambient air monitoring data in EPA's Air Quality
System (U.S. EPA, 2007) (http://www.epa.gov/ttn/airs/
airsaqs/). National and regional trends in this indicator are
based on the subset of NO2 monitoring stations that have
sufficient data to assess trends since 1980.
Exhibit 2-10. Ambient NC>2 concentrations in the
contiguous U.S. by EPA Region, 1980-2006ab
0.06
E
-S 0.05
,o
| 0.04
c
03
I 0.03
o
1 0.01
c
c
6=1 o.oo,
NAAQS = 0.053 ppm
-R1
R2
R3
R4
-R5
R6
R7
R8
-R9
R10b
-Nat'l
'82
'90 '92 '94 '96
Year
'02 '04 '06
EPA Regions
Coverage: 87 monitoring sites
in the EPA Regions (out of a
total of 369 sites measuring
I\I02 in 2006) that have
sufficient data to assess I\I02
trends since 1980.
bBecause N02 in Region 10 has
been at such low
concentrations, none of this
Region's monitoring sites have a complete record dating back to 1980.
Thus, no trend line for Region 10 is shown.
Data source: U.S. EPA, 2007
References
U.S. EPA (United States Environmental Protection
Agency). 2007. Data from the Air Quality System.
Accessed 2007.
U.S. EPA. 2003. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
U.S. EPA. 1995. Review of the national ambient air qual-
ity standards for nitrogen oxides: Assessment of scientific
and technical information. EPA/452/R-95/005. Research
Triangle Park, NC.
U.S. EPA. 1993. Air quality criteria for oxides of nitrogen.
EPA/600/8-91/049aF-cF. Research Triangle Park, NC.
14
-------
INDICATO
Volatile Organic Compounds Emissions
Volatile organic compounds (VOCs) are a large group of
organic chemicals that include any compound of carbon
(excluding carbon monoxide, carbon dioxide, carbonic acid,
metallic carbides or carbonates, and ammonium carbonate)
and that participate in atmospheric photochemical reac-
tions. VOCs are of interest in part because they contribute
to ozone formation (U.S. EPA, 2003a). Ozone (the Ozone
Concentrations indicator) is formed from chemical reactions
involving airborne VOCs, airborne nitrogen oxides, and
sunlight. VOCs are also of interest because many individual
VOCs are known to be harmful to human health (the
Benzene Concentrations indicator; the Air Toxics Emis-
sions indicator). Health effects vary by pollutant. VOCs are
emitted from a variety of sources, including motor vehicles,
chemical manufacturing facilities, refineries, factories,
consumer and commercial products, and natural (biogenic)
sources (mainly trees) (U.S. EPA, 2003b).
This indicator presents VOC emissions from tradition-
ally inventoried anthropogenic source categories:
(1) "Fuel combustion," which includes emissions from
coal-, gas-, and oil-fired power plants and industrial, com-
mercial, and institutional sources, as well as residential
heaters and boilers; (2) "Other industrial processes," which
includes chemical production, petroleum refining, metals
production, and processes other than fuel combustion; (3)
"On-road vehicles," which includes cars, trucks, buses, and
motorcycles; and (4) "Nonroad vehicles and engines," such
as farm and construction equipment, lawnmowers, chain-
saws, boats, ships, snowmobiles, aircraft, and others. The
indicator also includes estimates of biogenic VOC emis-
sions in 2002. Biogenic emissions were estimated using the
Biogenic Emissions Inventory System Model, Version 3.12,
•with data from the Biogenic Emissions Landcover Data-
base and 2001 annual meteorological data.
VOC emissions data are tracked by the National
Emissions Inventory (NEI). The NEI is a composite of
data from many different sources, including industry and
numerous state, tribal, and local agencies. Different data
sources use different data collection methods, and many of
the emissions data are based on estimates rather than actual
measurements. For most fuel combustion sources and
industrial sources, emissions are estimated using emission
factors. Emissions from on-road and nonroad sources were
estimated using EPA-approved modeling approaches (U.S.
EPA, 2007a).
NEI data have been collected since 1990 and cover all
50 states and their counties, D.C., the U.S. territories of
Puerto Rico and Virgin Islands, and some of the territories
of federally recognized American Indian nations. Data are
presented only for 1990 and the years from 1996 to 2002;
prior to 1996, only the 1990 data have been updated to be
comparable to the more recent inventories.
Exhibit 2-11. VOC emissions in the U.S. by
source category, 1990 and 1996-2002
25
IT 20
o
I 15
£ 10
_o
1 5
LLJ
0
A. Anthropogenic VOC emissions by source category3
-Mr
Fuel combustion
Other industrial processes
On-road vehicles
Nonroad vehicles and engines
'90
'96 '97
'99 '00 '01 '02
Year
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
Data source: U.S. EPA, 2007b
B. Relative amounts of VOC
emissions from anthropogenic
and biogenic sources, 2002
Anthropogenic
28%
>
Biogenic
72%
What the Data Show
This indicator focuses on trends in VOC emissions from
anthropogenic sources. However, VOC emissions from
biogenic sources were estimated for 2002 to provide a sense
of the relative contributions of natural versus anthropogenic
emissions. Nationally, biogenic emissions were estimated
to contribute approximately 72 percent to VOC emissions
from all sources during 2002 (Exhibit 2-11, panel B). Thus,
VOC emissions from biogenic sources are larger than the
VOC emissions from all anthropogenic sources combined.
According to NEI data, national total estimated
VOC emissions from anthropogenic sources, excluding
\vildfires and prescribed burns, decreased by 25 percent
between 1990 and 2002 (from 23,048,000 to 17,194,000
tons) (Exhibit 2-11, panel A). The over-whelming major-
ity of anthropogenic emissions reductions were observed
among industrial processes and on-road mobile sources.
Combined, these two source categories accounted for 84
percent of the total nation-wide estimated anthropogenic
VOC emissions in 1990 (excluding -wildfires and pre-
scribed burns), but accounted for only 72 percent of the
nation-wide anthropogenic emissions in 2002.
15
-------
INDICATO
Volatile Organic Compounds Emissions
Trends in estimated anthropogenic VOC emissions in
nine of the ten EPA Regions were consistent with the over-
all decline seen nationally from 1990 to 2002 (Exhibit 2-12).
Changes in VOC emissions ranged from a 52 percent reduc-
tion (Region 9) to a 16 percent increase (Region 10).
Indicator Limitations
• Comparable VOC emissions estimates through the NEI
are available only for 1990 and 1996-2002. Data for
1991-1995 are not provided due to differences in emis-
sions estimation methodologies from other inventory
years, which could lead to improper trend assessments.
• VOC emissions from "miscellaneous sources" are not
included in the total emissions. Details on emissions
from miscellaneous sources can be found by download-
ing 2002 NEI inventory data for the "nonpoint sector"
(http://www. epa.gov/ttn/chief/net/2002inventory.html).
• VOC emissions data are largely based on estimates that
employ emission factors generated from empirical and
engineering studies, rather than on actual measure-
ments of VOC emissions. These estimates are generated
using -well-established approaches, and quality assurance
measures are implemented to ensure that the emissions
data entered in NEI meet data quality standards (U.S.
EPA, 2006). Nonetheless, the estimates have uncertain-
ties inherent in the emission factors and emissions models
used to represent sources for which emissions have not
been directly measured.
• The methodology for estimating emissions is continually
reviewed and is subject to revision. Trend data prior to
any revisions must be considered in the context of
those changes.
• Not all states and local agencies provide the same data or
level of detail for a given year.
Data Sources
Summary data in this indicator were provided by EPA's
Office of Air Quality Planning and Standards, based on
biogenic and anthropogenic VOC emissions data in the
NEI (U.S. EPA, 2007b) (http://www.epa.gov/ttn/chief/
net/2002inventory.html). This indicator aggregates the
NEI data by source type (anthropogenic or biogenic),
source category, and EPA Region.
References
U.S. EPA (United States Environmental Protection
Agency). 2007a. Documentation for the final 2002 mobile
National Emissions Inventory, Version 3.
Exhibit 2-12. VOC emissions in the U.S. by
EPA Region, 1990 and 1996-20023
—R1
-R2
-R3
R4
-R5
R6
R7
R8
-R9
-R10
'96 '97 '98 '99 '00 '01 '02
Year
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
Data source: U.S. EPA, 2007b
U.S. EPA. 2007b. Data from the 2002 National Emissions
Inventory, Version 3.0. Accessed 2007.
U.S. EPA. 2006. NEI quality assurance and data augmenta-
tion for point sources. Research Triangle Park, NC.
U.S. EPA. 2003a. Requirements for preparation, adoption,
and submittal of implementation plans: Definitions. Code
of Federal Regulations 40CFR51.100(s).
U.S. EPA. 2003b. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
16
-------
Ambient Concentrations of Ozone
Ozone is a gas found in different parts of the atmosphere.
Ozone in the upper atmosphere, or stratosphere, helps
protect the Earth from the sun's harmful rays. (The Ozone
Levels over North America indicator describes trends in
stratospheric ozone levels over the U.S.) In the lowest level
of the atmosphere, the troposphere, ozone is harmful to
both human health and the environment. For this reason,
ozone is often described as being "good up high and bad
nearby" (U.S. EPA, 2003a). Although some industrial
sources release ozone directly into the environment, most
ground-level ozone forms in the air from chemical reac-
tions involving nitrogen oxides (NO ), volatile organic
compounds (VOCs), and sunlight. Ozone levels are typi-
cally highest during the afternoon hours of the summer
months, when the influence of direct sunlight is the great-
est. These highest levels occur during what is known as
the "ozone season," which typically occurs from May 1 to
September 30 but whose time frame varies by state (U.S.
EPA, 2003b).
Variations in weather conditions play an important role
in determining ozone levels. Daily temperatures, rela-
tive humidity, and wind speed can affect ozone levels. In
general, warm dry weather is more conducive to ozone
formation than cool wet weather. Wind can affect both
the location and concentration of ozone pollution. NO
and VOC emissions can travel hundreds of miles on air
currents, forming ozone far from the original emissions
sources. Ozone also can travel long distances, affecting
areas far downwind. High winds tend to disperse pol-
lutants and can dilute ozone concentrations. However,
stagnant conditions or light winds allow pollution levels to
build up and become more concentrated.
Inhalation exposure to ozone has been linked to numer-
ous respiratory health effects, including acute reversible
decrements in lung function, airway inflammation, cough,
and pain when taking a deep breath. Ozone exposure can
aggravate lung diseases such as asthma, leading to increased
medication use and increased hospital admission and visits
to emergency rooms. In addition, evidence is highly sug-
gestive that ozone directly or indirectly contributes to
non-accidental and cardiopulmonary-related mortality, but
the underlying mechanisms by which such effects occur
have not been fully established (U.S. EPA, 2006). Although
people with lung disease are most susceptible to the effects
of ozone, even healthy people who are active outdoors can
suffer from ozone-related health effects. Further, evidence
suggests that older adults (more than 65 years old) appear to
be at excess risk of ozone-related mortality or hospitaliza-
tion (U.S. EPA, 2006). Elevated concentrations of ozone
can also affect vegetation and ecosystems, as the Ozone
Injury to Forest Plants indicator describes further
(U.S. EPA, 2006).
Exhibit 2-13. Ambient 8-hour ozone
concentrations in the U.S., 1978-20063
:1°'14
; £ °'12
i re
; o
i ^ 0.10
; re
, -o
i £ 0 08
1 re
: g 0.06
• '-?
1 3
; g§ 0.04
| c
i o 0.02
: o
" 0.00
160
A. Ambient concentrations
,90% of sites have concentrations below this line
NAAQS = 0.08 ppm1
rwciayc ^i
Median*
10% of sites have concentrations below this line
•'88 '90-'92 '94-'96
Averaging period
B-'OO '02-'04
B. Number of trend sites above NAAQS
CO
; a
120
G o ^
HI
•
!§
o •
J3.E-S
E ! Jl
•? E s
40
0
78-
'82-'84 '86-'88 '90-'92 '94-'96 '98-'00 '02-'04
Averaging period
Coverage: 201 monitoring sites in 150 counties nationwide (out of
a total of 1,194 sites measuring ozone in 2006) that have sufficient
data to assess ozone trends since 1978.
bThe figure displays the 1997 NAAQS (0.08 ppm). Future versions of
the ROE will compare ozone concentrations to the recently
promulgated 2008 NAAQS (0.075 ppm) or to the NAAQS in effect at
the time.
Data source: U.S. EPA, 2007
This indicator presents ambient ground-level ozone
concentrations in parts per million (ppm) from 1978 to
2006. Data are shown for 8-hour averaging times, based
on continuous ozone monitoring data and consistent with
this pollutant's National Ambient Air Quality Standard
(NAAQS). The 8-hour standard is indicative of exposures
occurring over a sustained period of time (e.g., an outdoor
•worker's exposure over the course of a work day). Trends
for this indicator represent 201 sites in 150 counties nation-
wide that have data for the period of record in the State
and Local Air Monitoring Stations net-work or by other
special purpose monitors. The indicator also displays trends
17
-------
Ambient Concentrations of Ozone
Exhibit 2-14. Ambient 8-hour ozone
concentrations in the contiguous U.S. by EPA
Region, 1978-20063
0.16
i-'88 '94-'96 '02-'04
Averaging period
aCoverage:201 monitoring
sites in the EPA Regions (out of
a total of 1,194 sites measuring
ozone in 2006) that have
sufficient data to assess ozone
trends since 1978.
bThe figure displays the 1997
NAAQS (0.08 ppm). Future
versions of the ROE will
compare ozone concentrations
to the recently promulgated
2008 NAAQS (0.075 ppm) or
to the NAAQS in effect at
the time.
Data source: U.S. EPA, 2007
EPA Regions
nr «*
2^
in ozone measurements in each EPA Region. This indica-
tor's exhibits display the corresponding 1997 NAAQS as a
point of reference, but the fact that the national or regional
concentrations fall below the standard does not mean that
all monitoring sites nationally or in any EPA Region also
are below the standard. The indicator displays trends in
the number of the 201 sites nation-wide at which ozone
concentrations exceeded the level of the 1997 standard, but
this statistic is not displayed for each EPA Region.
Trends in ozone concentrations can be difficult to dis-
cern because of the year-to-year variations in the concen-
trations. By presenting data for rolling 3-year time periods,
this indicator smoothes out the "peaks" and "valleys" in
the trend, making it easier to see the long-term trend.
Three years is consistent with the 3-year period used to
assess compliance with the ozone standards. For the 8-hour
trends in this report, a 3-year average of the fourth highest
daily maximum 8-hour concentration in each year is used
to be consistent with the 8-hour ozone standard.
What the Data Show
Between the 1978-1980 and 2004-2006 averaging periods,
nation-wide fourth highest daily maximum 8-hour ambi-
ent ozone concentrations decreased by 25 percent (Exhibit
2-13, panel A). Although the 8-hour ozone levels in
2004-2006 -were the lowest on record and the number of
trend sites measuring ozone concentrations above the level
of the 1997 8-hour NAAQS decreased by 75 percent over
the time frame covered in this indicator (Exhibit 2-13,
panel B), ambient air monitoring data collected in 2006
and reported to EPA's Air Quality System indicate that
approximately 77 million people lived in counties -where
8-hour average ozone concentrations are above the level
of the 1997 primary ozone NAAQS. Among the ten EPA
Regions, the most substantial declines in 8 hour levels -were
observed in EPA Regions that originally had the high-
est ozone concentrations (EPA Regions 1 and 9) (Exhibit
2-14). Over the entire period of record, Region 10 consis-
tently showed the lowest Regional ozone levels.
Also shown in Exhibit 2-13 (panel A) are the 90th and
10th percentiles based on the distribution of statistics at
the monitoring sites. This provides additional graphical
representation of the variability of measured concentrations
across the monitoring sites for a given 3-year period. Thus,
the graphic displays the concentration range -where 80 per-
cent of measured values occurred for that 3-year period.
In summary, despite reductions in ambient concentra-
tions of ozone over the past quarter century and decreases
in the emissions of ozone precursors since 1990 (the
Nitrogen Oxides Emissions indicator; the VOC Emissions
indicator), ozone remains one of the most persistent and
ubiquitous air pollution issues in the U.S.
Indicator Limitations
• Short-term trends in ozone concentrations are often
highly dependent on meteorological conditions. This
complicates efforts to interpret data for any given year.
Air quality trends over the longer term are far less likely
to be influenced by unusual meteorological conditions.
• Because most of the monitoring sites are located in urban
areas, the trends might not accurately reflect conditions
outside the immediate urban monitoring areas.
• Because of the relatively small number of trend sites in
some EPA Regions, the regional trends are subject to
greater uncertainty than the national trends. Some EPA
Regions -with low average concentrations may include
areas -with high local concentrations, and vice versa.
18
-------
Ambient Concentrations of Ozone
• To ensure that long-term trends are based on a consistent
set of monitoring sites, selection criteria were applied to
identify the subset of ozone monitoring sites with sufficient
data to assess trends since 1978. Monitoring sites without
sufficient data are not included in the trend analysis. Some
excluded monitoring sites reported ozone concentrations
above the level of the ozone standard over the time frame
covered by this indicator. In 2006, for example, 187 moni-
toring sites (in addition to the trend sites shown in Exhibit
2-13, panel B) recorded ozone concentrations above the
level of the 1997 NAAQS, but did not have sufficient long-
term data to be included in this indicator.
Data Sources
Summary data in this indicator were provided by EPA's
Office of Air Quality Planning and Standards, based on
ozone ambient air monitoring data in EPA's Air Quality
System (U.S. EPA, 2007) (http://www.epa.gov/ttn/airs/
airsaqs/). National and regional trends in this indicator are
based on the subset of ozone monitoring stations that have
sufficient data to assess trends since 1978.
References
U.S. EPA (United States Environmental Protection
Agency). 2007. Data from the Air Quality System.
Accessed 2007.
U.S. EPA. 2006. Air quality criteria for ozone and related
photochemical oxidants. EPA/600/R-05/004aF-cF.
Research Triangle Park, NC.
U.S. EPA. 2003a. Ozone: Good up high, bad nearby.
EPA/451/K-03/001. Washington, DC.
U.S. EPA. 2003b. Latest findings on national air qual-
ity—2002 status and trends. EPA/454/K-03/001. Research
Triangle Park, NC.
19
-------
INDICATO
Ozone Injury to Forest Plants
Air pollution can have note-worthy cumulative impacts
on forested ecosystems by affecting regeneration,
productivity, and species composition (U.S. EPA, 2006). In
the U.S., ozone in the lower atmosphere is one of the pol-
lutants of primary concern. Ozone injury to forest plants
can be diagnosed by examination of plant leaves. Foliar
injury is usually the first visible sign of injury to plants
from ozone exposure and indicates impaired physiological
processes in the leaves (Grulke, 2003).
This indicator is based on data from the U.S. Depart-
ment of Agriculture (USDA) Forest Service Forest Inven-
tory and Analysis (FIA) program. As part of its Phase 3
program, formerly known as Forest Health Monitoring,
FIA examines ozone injury to ozone-sensitive plant species
at ground monitoring sites in forest land across the coun-
try. For this indicator, forest land does not include -wood-
lots and urban trees. Sites are selected using a systematic
sampling grid, based on a global sampling design (White
et al, 1992; Smith et al, 2003). At each site that has at least
30 individual plants of at least three ozone-sensitive species
and enough open space to ensure that sensitive plants are
not protected from exposure by the forest canopy, FIA
looks for damage on the foliage of ozone-sensitive forest
plant species. Because ozone injury is cumulative over the
course of the growing season, examinations are conducted
in July and August, -when ozone injury is typically highest.
Monitoring of ozone injury to plants by the USDA For-
est Service has expanded over the last 10 years from moni-
toring sites in ten states in 1994 to nearly 1,000 monitoring
sites in 41 states in 2002. The data underlying this indica-
tor are based on averages of all observations collected in
2002, the latest year for -which data are publicly available,
and are broken down by EPA Region. Ozone damage to
forest plants is classified using a subjective five-category
biosite index based on expert opinion, but designed to
be equivalent from site to site. Ranges of biosite values
translate to no injury, low or moderate foliar injury (visible
foliar injury to highly sensitive or moderately sensitive
plants, respectively), and high or severe foliar injury, -which
•would be expected to result in tree-level or ecosystem-
level responses, respectively (Coulston et al., 2004; U.S.
EPA, 2006).
What the Data Show
There is considerable regional variation in ozone injury
to sensitive plants (Exhibit 2-15). The highest percent-
ages of observed high and severe foliar injury, -which are
most likely to be associated -with tree or ecosystem-level
responses, are primarily found in the Mid-Atlantic and
Southeast regions. In EPA Region 3, 12 percent of ozone-
sensitive plants showed signs of high or severe foliar dam-
age, and in Regions 2 and 4, the values -were 10 percent
and 7 percent, respectively. The sum of high and severe
ozone injury ranged from 2 percent to 4 percent in EPA
Exhibit :
the U.S.
Region 1
(54 sites)
Region 2
(42 sites)
Region 3
(111 sites)
Region 4
(227 sites)
Region 5
(180 sites)
Region 6
(59 sites)
Region 7
(63 sites)
Region 8
(72 sites)
Region 9
(80 sites)
Region 10
(57 sites)
M5. Ozone injury to forest plants in
by EPA Region, 2002ab
Degree of injury:
None
Low Moderate High Severe
3ercenr. of monitoring sites in each category:
68.5 16.7 11.1
-3.7
61.9 21.4 7.1 7
55.9 18.0 14.4 7
75.3 10.1 7.0 "
|2.4
|4.5
13.5
-4.0
75.6 18.3 6
.1
94.9
-5.1
85.7 9.5
-3.2
•1.6
100.0
76.3 12.5 8.8
1.3
•1.3
100.0
Coverage: 945 monitc
located in 41 states.
bTotals may not add to
rounding.
Data source: USDA Ft
2006
.ring sites, EPA Re9'°ns
-------
INDICATOR
Ozone Injury to Forest Plants
• Because species distributions vary regionally, different
ozone-sensitive plant species were examined in different
parts of the country. These target species could vary with
respect to ozone sensitivity, which might account for
some of the apparent differences in ozone injury among
EPA Regions.
• Ozone damage to foliage is considerably reduced under
conditions of low soil moisture, but most of the vari-
ability in the index (70 percent) was explained by ozone
concentration (Smith et al., 2003).
• Ozone may have other adverse impacts on plants (e.g.,
reduced productivity) that do not show signs of visible
foliar injury (U.S. EPA, 2006).
• Though FIA has extensive spatial coverage based on a
robust sample design, not all forested areas in the U.S. are
monitored for ozone injury.
• Even though the biosite data have been collected over
multiple years, most biosites were not monitored over the
entire period, so these data cannot provide more than a
baseline for future trends.
Data Sources
Data were provided by the USDA Forest Service's Ozone
Biomonitoring Program, which maintains a database of
plant injury statistics by state (USDA Forest Service, 2006)
(http://nrs.fs.fed.us/fia/topics/ozone/data/). This indicator
aggregates the state data by EPA Region.
References
Coulston, J.W., K.H. Riitters, and G.C. Smith. 2004. A
preliminary assessment of the Montreal process indica-
tors of air pollution for the United States. Environ. Monit.
Assess. 95:57-74.
Grulke, N.E. 2003. The physiological basis of ozone injury
assessment attributes in Sierran conifers. In: Bytnerowicz,
A., MJ. Arbaugh, and R. Alonso, eds. Ozone air pollution
in the Sierra Nevada: Distribution and effects on forests.
New York, NY: Elsevier Science, Ltd. pp. 55-81.
Smith, G., J. Coulston, E. Jepsen, and T. Prichard. 2003. A
national ozone biomonitoring program—results from field
surveys of ozone sensitive plants in Northeastern forests
(1994-2000). Environ. Monit. Assess. 87:271-291.
USDA Forest Service (United States Department of Agri-
culture Forest Service). 2006. Ozone bioindicator data.
Accessed 2006.
U.S. EPA (United States Environmental Protection Agency).
2006. Air quality criteria for ozone and related photochemi-
cal oxidants. EPA/600/R-05/004aF-cF. Research Triangle
Park, NC.
White, D., AJ. Kimerlmg, and W.S. Overton. 1992. Car-
tographic and geometric component of a global sampling
design for environmental monitoring. Cartogr. Geograph.
Info. Sys. 19:5-22.
21
-------
Particulate Matter Emissions
u
Particulate matter" (PM) is the general term used to
describe solid particles and liquid droplets found in the
air. The composition and size of these airborne particles
and droplets vary. Some particles are large enough to be
seen as dust or dirt, while others are so small they can only
be seen using a powerful microscope. Two size ranges,
known as PM1Q and PM2 5, are widely monitored, both at
major emissions sources and in ambient air. PM1Q includes
particles that have aerodynamic diameters less than or
equal to 10 microns ((am), approximately equal to one-
seventh the diameter of human hair. PM2 5 is the subset of
PM1Q particles that have aerodynamic diameters less than
or equal to 2.5 (am.
Particles within the two size ranges behave differently in
the atmosphere. PM2 5, or fine particles, can remain air-
borne for long periods and travel hundreds of miles. Coarse
particles, or the subset of PM1Q that is larger than 2.5 (am,
do not remain airborne as long and their spatial impact is
typically limited because they tend to deposit on the ground
downwind of emissions sources. Larger coarse particles are
not readily transported across urban or broader areas because
they are generally too large to follow air streams and they
tend to be removed easily on contact with surfaces. In short,
as the particle size increases, the amount of time the particles
remain airborne decreases. The PM Concentrations indica-
tor describes the various ways PM can harm human health
and the environment (U.S. EPA, 2004).
PM can be emitted directly or formed in the atmo-
sphere. "Primary" particles are those released directly to
the atmosphere. These include dust from roads and soot
from combustion sources. In general, coarse PM is com-
posed largely of primary particles. "Secondary" particles,
on the other hand, are formed in the atmosphere from
chemical reactions involving primary gaseous emissions.
Thus, these particles can form at locations distant from
the sources that release the precursor gases. Examples
include sulfates formed from sulfur dioxide emissions from
power plants and industrial facilities and nitrates formed
from nitrogen oxides released from power plants, mobile
sources, and other combustion sources. Unlike coarse PM,
a much greater portion of fine PM (PM2 5) contains sec-
ondary particles (U.S. EPA, 2004).
This indicator presents trends in annual average pri-
mary PM emissions data tracked by the National Emis-
sions Inventory (NEI). The NEI tracks emission rate data,
both measured and estimated, for primary particles only.
Because secondary particles are not released directly from
stacks, the NEI instead tracks the precursors that contrib-
ute to formation of secondary particles. These precursors
include nitrogen oxides, sulfur dioxide, ammonia, and
other gases (e.g., particle-producing organic gases), some
of which are addressed in separate indicators (the Nitrogen
Oxides Emissions indicator; the Sulfur Dioxide Emissions
Exhibit 2-16. PM-io emissions in the U.S. by
source category, 1990 and 1996-2002
3.5
-. 3.°
tfi
S 2.5
c
o
= 2.0
_£_
« 1.5
1 1.0
E
m 0.5
0.0
A. Anthropogenic PM10 emissions by source category3
Fuel combustion
Other industrial processes
On-road vehicles
Nonroad vehicles and engines
'90
'96 '97
'99
'00
'01
'02
Year
B. Relative amounts of PM10
emissions from anthropogenic
and other sources, 2002b
Miscellaneous
and natural
sources
26%
Anthropogenic
14%
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years
are fully up to date. Data are
available for inventory years
1991-1995, but these data
have not been updated to
allow comparison with data
from 1990 and 1996-2002.
Starting in 1999, EPA began
tracking condensable
particulate emissions
separately from filterable
particulate emissions. In
order to display data
generated using a consistent methodology, emissions of
condensable particulate from 1990 to 2002 are not included in
Panel A. However, condensable particulate emissions are
included in Panel B.
Data source: U.S. EPA, 2007b
x
Fugitive dust
60%
indicator). Particles formed through secondary processes
are not included in this indicator.
Primary emissions of PM can exist as solid or liquid
matter (the "filterable" portion) or as gases (the "condens-
able" portion). Data for the condensable portion exist only
for the years 1999 to 2002. To allow for a valid comparison
of emissions trends from 1990 to 2002, only data for the
filterable portion of PM1Q and PM2 5 are included in the
trend graphs. Condensables are, however, included in the
inset pie charts shown in Exhibits 2-16 and 2-18 (i.e., panel
B in both exhibits).
All emissions data presented in this indicator are taken
from the NEI. Primary particulate emissions data are pre-
sented for the traditionally inventoried anthropogenic source
categories: (1) "Fuel combustion," which includes emissions
from coal-, gas-, and oil-fired power plants and industrial,
22
-------
Particulate Matter Emissions
Exhibit 2-17. PM-m emissions in the U.S. by
EPA Region, 1990 and 1996-2002ab
'90 '96 '97 '98 '99 '00 '01 '02
Year
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
fully up to date. Data are
available for inventory years
1991-1995, but these data
have not been updated to allow
comparison with data from
1990 and 1996-2002.
Starting in 1999, EPA began tracking condensable particulate
emissions separately from filterable particulate emissions. In
order to display data generated using a consistent methodology,
emissions of condensable particulate from 1999 to 2002 are not
included in this figure.
Data source: U.S. EPA, 2007b
commercial, and institutional sources, as well as residential
heaters and boilers; (2) "Other industrial processes," which
includes chemical production, petroleum refining, met-
als production, and processes other than fuel combustion;
(3) "On-road vehicles," which includes cars, trucks, buses,
and motorcycles; and (4) "Nonroad vehicles and engines,"
such as farm and construction equipment, lawnmowers,
chainsaws, boats, ships, snowmobiles, aircraft, and others.
For 2002 only, this indicator includes a comparison of these
anthropogenic sources with emissions from miscellaneous
and natural sources, such as agriculture and forestry, -wild-
fires and managed burning, and fugitive dust from paved and
unpaved roads. Biogenic emissions were estimated using the
Biogenic Emissions Inventory System Model, Version 3.12,
•with data from the Biogenic Emissions Landcover Database
and 2001 annual meteorological data. The NEI also docu-
ments estimates of primary emissions from fugitive dust and
miscellaneous sources.
The NEI is a composite of data from many different
sources, including industry and numerous state, tribal, and
local agencies. Different data sources use different data
collection methods, and many of the emissions data are
based on estimates rather than actual measurements. For
most fuel combustion sources and industrial sources, emis-
sions are estimated using emission factors. Emissions from
on-road and nonroad sources were estimated using EPA-
approved modeling approaches (U.S. EPA, 2007a).
NEI data have been collected since 1990 and cover all
50 states and their counties, D.C., the U.S. territories of
Puerto Rico and Virgin Islands, and some of the territories
of federally recognized American Indian nations. Data are
presented for 1990 and the years from 1996 to 2002; prior
to 1996, only the 1990 data have been updated to be com-
parable to the more recent inventories.
What the Data Show
Primary PM10 Emissions Trends
Estimated primary PM1Q emissions from anthropogenic
sources decreased 27 percent nationally between 1990 and
2002 (Exhibit 2-16, panel A). Of these sources, those in
the fuel combustion category saw the largest absolute and
relative decrease in emissions (656,000 tons; 55 percent).
Primary PM1Q emissions from the group of sources includ-
ing miscellaneous and natural sources and fugitive dust
•were estimated to account for 86 percent of total primary
PM1Q emissions (including condensables from stationary and
mobile sources) in 2002, the majority of which was attribut-
able to fugitive dust from roads (Exhibit 2-16, panel B).
Changes in estimated primary anthropogenic PM1Q
emissions from 1990 to 2002 varied widely among EPA
Regions, ranging from an increase of 16 percent (Region
8) to a decrease of 75 percent (Region 2) (Exhibit 2-17).
Primary PM25 Emissions Trends
Estimated primary PM2 5 emissions from anthropogenic
sources decreased 44 percent nationally between 1990 and
2002 (Exhibit 2-18, panel A). The largest absolute and
relative decline in PM2 was seen in the fuel combustion
source category (621,000 tons; 68 percent). Primary emis-
sions from the group of sources including miscellaneous
and natural sources and fugitive dust were estimated to
account for 64 percent of the total PM2 5 emissions (includ-
ing condensables from stationary and mobile sources)
nationally in 2002 (Exhibit 2-18, panel B).
Primary anthropogenic PM2 5 emissions decreased in all
ten EPA Regions from 1990 to 2002, with percent reduc-
tions ranging from 21 percent (Region 4) to 71 percent
(Region 2) (Exhibit 2-19).
Indicator Limitations
• Comparable PM emissions estimates through the NEI
are available only for 1990 and 1996-2002. Data for
1991-1995 are not provided due to differences in emis-
sions estimation methodologies from other inventory
years, which could lead to improper trend assessments.
23
-------
Particulate Matter Emissions
Exhibit 2-18. PM2.5 emissions in the U.S. by
source category, 1990 and 1996-2002
2.5
? 2.0
_o
I 1.5
A. Anthropogenic PM2.5 emissions by source category
J-V
Other industrial processes
On-road vehicles
Nonroad vehicles and engines
'90
'96
'97
'99
'01
'02
Year
B. Relative amounts of PM2.5
emissions from anthropogenic
and other sources, 2002b
Miscellaneous
and natural
sources
33%
Anthropogenic
36%
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years
are fully up to date. Data are
available for inventory years
1991-1995, but these data
have not been updated to
allow comparison with data
from 1990 and 1996-2002.
Starting in 1999, EPA began
tracking condensable
particulate emissions
separately from filterable
particulate emissions. In
order to display data
generated using a consistent methodology, emissions of
condensable particulate from 1990 to 2002 are not included in
Panel A. However, condensable particulate emissions are included
in Panel B.
Data source: U.S. EPA, 2007b
3%
Fugitive dust
31%
Because the emissions indicators focus on sources of
anthropogenic origin, PM emissions from miscella-
neous sources (e.g., wildfires) are not included in the
trend line. Details on emissions from these sources can
be found by downloading 2002 NEI inventory data for
the "nonpoint sector" (http://www.epa.gov/ttn/chief/
net/2002inventory.html).
The emissions data for PM are largely based on estimates
that employ emission factors generated from empirical
and engineering studies, rather than on actual measure-
ments of PM emissions. Although these estimates are
generated using -well-established approaches, the esti-
mates have uncertainties inherent in the emission factors
and emissions models used to represent sources for which
emissions have not been directly measured.
Exhibit 2-19. PM2 5 emissions in the U.S. by
EPA Region, 1990 and 1996-2002ab
'96 '97 '98 '99 '00 '01
Year
'02
aData are presented for 1990 ^ EPA Regions
and 1996-2002, as datasets
from these inventory years are
fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
Starting in 1999, EPA began tracking condensable particulate
emissions separately from filterable particulate emissions. In
order to display data generated using a consistent methodology,
emissions of condensable particulate from 1999 to 2002 are not
included in this figure.
Data source: U.S. EPA, 2007b
' The methodology for estimating emissions is continu-
ally reviewed and is subject to revision. Trend data prior
to these revisions must be considered in the context of
those changes.
• The indicator tracks primary PM emissions. Particles
that form in the air through secondary processes are not
included in this indicator, but are considered in the PM
Concentrations indicator.
• Not all states and local agencies provide the same data or
level of detail for a given year.
Data Sources
Summary data in this indicator were provided by EPA's
Office of Air Quality Planning and Standards, based on
biogenic and anthropogenic PM emissions data in the
NEI (U.S. EPA, 2007b) (http://www.epa.gov/ttn/chief/
net/2002inventory.html). This indicator aggregates the
NEI data by source type (anthropogenic or biogenic),
source category, and EPA Region.
24
-------
Particulate Matter Emissions
References
U.S. EPA (United States Environmental Protection
Agency). 2007a. Documentation for the final 2002 mobile
National Emissions Inventory, Version 3.
U.S. EPA. 2007b. Data from the 2002 National Emissions
Inventory, Version 3.0. Accessed 2007.
U.S. EPA. 2004. Air quality criteria for particulate mat-
ter (October 2004). EPA 600/P-99/002aF-bF. Research
Triangle Park, NC.
25
-------
Ambient Concentrations of Particulate Matter
F^articulate matter" (PM) is the general term used for a
mixture of solid particles and liquid droplets found in
the air. Airborne PM comes from many different sources.
"Primary" particles are released directly into the atmo-
sphere from sources such as cars, trucks, heavy equipment,
forest fires, and burning waste. Primary particles also
consist of crustal material from sources such as unpaved
roads, stone crushing, construction sites, and metallurgi-
cal operations. "Secondary" particles are formed in the air
from reactions involving precursor chemicals such as sul-
fates (which are formed from sulfur dioxide emissions from
power plants and industrial facilities), nitrates (which are
formed from nitrogen dioxide emissions from cars, trucks,
and power plants), and carbon-containing reactive organic
gas emissions from cars, trucks, industrial facilities, forest
fires, and biogenic sources such as trees.
Ambient air monitoring stations throughout the coun-
try measure air concentrations of two size ranges of parti-
cles: PM25 and PM1Q. PM25 consists of "fine particles" with
aerodynamic diameters less than or equal to 2.5 microns
(jam). PM1Q includes both fine particles (PM25) and "coarse
particles," which is the subset of PM1Q that is larger than
2.5 (am and smaller than 10 (am. The chemical makeup of
particles varies across the U.S. For example, fine particles
in the eastern half of the U.S contain more sulfates than
those in the West, while fine particles in southern Califor-
nia contain more nitrates than those in other areas of the
U.S. Carbon is a substantial component of fine particles
everywhere (U.S. EPA, 2004a).
Fine particles also have seasonal patterns. PM2 5 values
in the eastern half of the U.S. are typically higher in the
third calendar quarter (July-September), when sulfates are
more commonly formed from sulfur dioxide emissions
from power plants in that part of the country. Fine particle
concentrations tend to be higher in the fourth calendar
quarter (October-December) in many areas of the West, in
part because fine particle nitrates are more readily formed
in cooler weather, and wood stove and fireplace use pro-
duces more carbon.
Many recent epidemiologic studies show statistically
significant associations of various ambient PM indica-
tors (e.g., coarse or fine particulate, short-term or long-
term concentrations) with a variety of cardiovascular and
respiratory health endpoints, including mortality, hospital
admissions, emergency department visits, other medical
visits, respiratory illness and symptoms, and physiologic
changes in pulmonary function (U.S. EPA, 2004b). Sensi-
tive groups that appear to be at greatest risk to such PM
effects include older adults, individuals with cardiopulmo-
nary disease such as asthma or congestive heart disease, and
children (U.S. EPA, 2004b). Unlike other criteria pollut-
ants, PM is not a single specific chemical entity, but rather
a mixture of particles from different sources with different
Exhibit 2-20. Ambient 24-hour
concentrations in the U.S., 1988-20063
A. Ambient concentrations
NAAQS = 150ug/m3
90% of sites have concentrations below this line
10% of sites have
concentrations below this line
'90 '92 '94 '96
'00 '02 '04 '06
Year
B. Number of trend sites above NAAQS
'00 '02 '04 '06
Year
Coverage: 301 monitoring sites in 199 counties nationwide (out of a
total of 902 sites measuring PM10 in 2006) that have sufficient data
to assess PM-|0 trends since 1988.
Data source: U.S. EPA, 2007
sizes and chemical compositions, lexicological studies
suggest that some airborne particles are more toxic than
others, due to differences in their chemical composition—
a topic that is thoroughly reviewed in other publications
(e.g., U.S. EPA, 2004b).
PM also can cause adverse impacts to the environment.
Fine particles are the major cause of reduced visibility in
parts of the U.S., including many National Parks and Wil-
derness Areas (the Regional Haze indicator). PM deposi-
tion affects vegetation and ecosystems by altering nutrient
and chemical cycles in soils and surface water. For exam-
ple, deposition of particles containing nitrogen and sulfur
may change the nutrient balance and acidity of aquatic
environments so that species composition and buffering
capacity change (the Lake and Stream Acidity indicator).
Some particles that deposit onto plant leaves can corrode
26
-------
Ambient Concentrations of Particulate Matter
Exhibit 2-21. Ambient 24-hour
concentrations in the contiguous U.S. by EPA
Region, 1988-2006a
180
160
i — 140
: 1,120
i _=;
: = 100
' .2
? 2 80
! § 60
| C
; ° 40
20
Oi
NAAQS = 150 ug/m3
R1
R2
R3
R4
-R5
R6
R7
R8
-R9
-R10
-Nat'l
'90 '92 '94 '96 '9
Year
Coverage: 292 monitoring sites
in the EPA Regions (out of a total
of 902 sites measuring PM-|0 in
2006) that have sufficient data to
assess PM10 trends since 1988.
Data source: U.S. EPA, 2007
'00 '02 '04 '06
EPA Regions
leaf surfaces or interfere with plant metabolism. PM also
causes soiling and erosion damage to materials, including
monuments, statues, and other objects of cultural impor-
tance (U.S. EPA, 2004b).
This indicator presents trends in PM1Q and PM2 B
concentrations, using averaging times consistent with the
pollutants' corresponding National Ambient Air Quality
Standards (NAAQS). For PMW, trend data from 1988 to
2006 are presented for the second highest 24-hour con-
centrations measured at the trend sites during each cal-
endar year. For PM2 B, trend data from 1999 to 2006 are
presented for seasonally -weighted annual average concen-
trations and for the 98th percentiles of 24-hour average
concentrations measured at the trend sites over three con-
secutive calendar years. Trend data are based on measure-
ments from the State and Local Air Monitoring Stations
net-work and from other special purpose monitors. This
indicator presents PM1Q trends for 301 monitoring sites in
199 counties nation-wide and PM2 5 trends for 752 monitor-
ing sites in 508 counties nation-wide. For both PM1Q and
PM2 , the indicator displays trends for the entire nation
and for the ten EPA Regions.
The indicator's exhibits display the pollutants' NAAQS as
points of reference. However, the fact that the national val-
ues or those shown for EPA Regions fall below the standards
does not mean that all monitoring sites nationally or in any
particular EPA Region also are below the standards. The
Exhibit 2-22. Ambient annual PIVb.s concentrations
in the U.S., 1999-2006a
20
15
10
A. Ambient concentrations
90% of sites have concentrations below this line
»
NAAQS =15 ug/m3
Median
A
iverage
10% of sites have concentrations below this line
o
'99-'01
'00-'02
'01-'03 '02-'04
Averaging period
'03-'05
'04-'06
B. Number of trend sites above NAAQS
99-'01 '00-'02 '01-'03 '02-'04 '03-'05 '04-'06
Averaging period
Coverage: 752 monitoring sites in 508 counties nationwide (out of a
total of 786 sites measuring PM2.5 in 2006) that have sufficient data
to assess PM2.5 trends since 1999.
Data source: U.S. EPA, 2007
indicator displays trends in the number of PM1Q monitoring
sites and PM2 B monitoring sites nation-wide that recorded
ambient air concentrations above the level of the standards,
but these statistics are not displayed for each EPA Region.
What the Data Show
PM10 Concentration Trends
In 2006, the national 24-hour PM1Q concentration (based
on the second highest 24-hour concentration at each site)
•was 37 percent lower than the average 1988 level (Exhibit
2-20, panel A). Additionally, of the 301 sites used to
determine this trend (out of 902 total monitoring sites
that -were operating in 2006), the number reporting PM1Q
concentrations above the level of the 24-hour standard
declined 78 percent between 1988 and 2006 (Exhibit
2-20, panel B). All EPA Regions experienced a steady
27
-------
Ambient Concentrations of Particulate Matter
Exhibit 2-23. Ambient annual PIVb.s concentrations
in the contiguous U.S. by EPA Region, 1999-20063
20
ro £~15
I g-g
= O O)
= CD 3:
ro CD
^^e
•a r s 10
fg|
^ "ro _S2
1 i s 5
o ^
S "
Sd
o
CO U
l\IAAQS = 15ug/m3
0
'99-'01 '00-'02
-R1
R2
R3
R4
-R5
R6
R7
R8
-R9
-R10
-Nat'l
'01-'03 '02-'04
Averaging period
'03-'05 '04-'06
Coverage: 736 monitoring
sites in the EPA Regions (out
of a total of 786 sites
measuring PM2.5 in 2006) that
have sufficient data to assess
PM2.5 trends since 1999.
Data source: U.S. EPA, 2007
EPA Regions
decrease in 24-hour PM1Q levels over this period (Exhibit
2-21). EPA Region 10 showed the greatest relative
decrease (68 percent) since 1988.
Also shown in Exhibit 2-20 (panel A) are the 90th and
10th percentiles based on the distribution of annual statistics
at the monitoring sites. This provides additional graphical
representation of the distribution of measured concentra-
tions across the monitoring sites for a given year. Thus, the
graphic displays the concentration range where 80 percent
of measured values occurred for that year. (Note that this
presentation style also applies to panel A in Exhibits 2-22
and 2-24, discussed below.)
PM25 Concentration Trends
Seasonally -weighted average PM2 concentrations over the
2004-2006 averaging period were the lowest since nation-
wide monitoring began in 1999 (Exhibit 2-22, panel A).
The trend is based on measurements collected at 752 moni-
toring stations that have sufficient data to assess trends over
that period. The seasonally -weighted annual average con-
centrations decreased 10 percent between the 1999-2001
averaging period and the 2004-2006 averaging period.
The number of monitoring sites in this trend (752 out
of 786 total sites that -were operating in 2006) reporting
ambient air concentrations above the level of the annual
average PM2 B standard declined 61 percent over this period
(Exhibit 2-22, panel B).
Exhibit 2-24. Ambient 24-hour
concentrations in the U.S., 1999-20063
60
45
A. Ambient concentrations
_2 — re
•=
- -I—' QJ
•^ s
on
30
15
90% of sites have
concentrations below this line
NAAQS = 35 ug/m;
10% of sites have
concentrations below this line
'99-'01
'00-'02
'01-'03 '02-'04
Averaging period
'03-'05
'04-'06
B. Number of trend sites above NAAQS
99-'01 '00-'02 '01-'03 '02-'04 '03-'05 '04-'06
5 o
'- ° Averaging period
Coverage: 752 monitoring sites in 508 counties nationwide (out of a
total of 811 sites measuring PM2.5 in 2006) that have sufficient data
to assess PM25 trends since 1999.
Data source: U.S. EPA, 2007
Regional declines -were greatest in portions of the West
(EPA Region 9), the Southeast (EPA Region 4), and the
Mid-west (EPA Region 5), -where seasonally -weighted aver-
age PM2 B levels over the 2004-2006 averaging period -were
19 percent, 11 percent, and 11 percent lower than those in
1999-2001 averaging period, respectively (Exhibit 2-23).
In 2004-2006, the average of 98th percentiles of 24-hour
PM2 concentrations at the 752 monitoring sites used
for the trend was 10 percent lower than the 1999-2001
level (Exhibit 2-24, panel A). The number of monitoring
sites in this trend (752 out of a total of 811 sites that -were
operating in 2006) reporting ambient air concentrations
above the level of the 24-hour PM2 standard declined
46 percent over this period (Exhibit 2-24, panel B). All
ten EPA Regions experienced decreasing 24-hour PM2 5
levels between the 1999-2001 averaging period and the
28
-------
Ambient Concentrations of Particulate Matter
Exhibit 2-25. Ambient 24-hour
concentrations in the contiguous U.S. by EPA
Region, 1999-2006a
• 60
O •
O »-
o ^
1_ "O
45
-i 30
o
— CD -1 R
•4= c/> I 0
££ o
U.S. EPA. 2004a. The particle pollution report: Cur-
rent understanding of air quality and emissions through
2003. EPA 454/R-04/002. Research Triangle Park, NC.
U.S. EPA. 2004b. Air quality criteria for particulate mat-
ter (October 2004). EPA 600/P-99/002aF-bF. Research
Triangle Park, NC.
29
-------
INDICATO
Regional Haze
Visibility impairment occurs when air pollution, especially
particles, scatter and absorb light. The resulting haze not
only limits the distance one can see, but also degrades the
color, clarity, and contrast of scenes. As the PM Concentra-
tions indicator describes further, the same pollutants that
impair visibility are linked to serious health effects. Visibility
impairment occurs throughout the country, including both
urban and rural areas. Regional haze is visibility impair-
ment caused by the cumulative air pollutant emissions from
numerous sources over a wide geographic area (U.S. EPA,
2004a). Regional haze has been identified as an important
issue for all of the National Parks and Wilderness Areas,
such as the Grand Canyon, Great Smoky Mountains, Mount
Rainier, Shenandoah, Yellowstone, and Yosemite National
Parks (U.S. EPA, 2003).
The particles that impair visibility include both primary
and secondary pollutants. The primary pollutants of con-
cern are particles that are emitted directly into the atmo-
sphere, such as dust from roads or soot (elemental carbon)
from combustion sources (e.g., wood combustion). Sec-
ondary pollutants of concern are particles that form in the
atmosphere from chemical reactions and physical processes,
such as sulfates (formed from sulfur dioxide emissions from
power plants and other industrial facilities) and nitrates
(formed from nitrogen oxides emitted from power plants,
automobiles, and other types of combustion sources).
Humidity can increase the effect of pollution on vis-
ibility, causing some particles to become more efficient
at scattering light and impairing visibility (U.S. EPA,
2003). In the eastern U.S., where annual average relative
humidity levels are between 70 percent and 80 percent,
reduced visibility mainly results from secondarily formed
sulfates and high humidity, along with a somewhat lower
contribution from organic carbon and nitrates (U.S. EPA,
2004b). The effect of humidity is particularly strong
in summer. Humidity is less of a factor in the West, as
average values are generally between 50 percent and 60
percent. In western states, primary emissions from sources
like wood smoke and nitrates contribute a large percentage
of the total particulate loading, though secondarily formed
sulfates also contribute to visibility impairment. With-
out the effects of anthropogenic sources of pollution, the
annual average natural visual range in the U.S. would vary
•with location, and is estimated to range from 75 to 150 km
(45 to 90 miles) in the East and from 200 to 300 km (120
to 180 miles) in the West (U.S. EPA, 2003).
This indicator reports visibility estimates calculated
from measurements of particulate matter (PM) constituents
collected at 38 monitoring sites between 1992 and 2004
at National Parks, Wilderness Areas, and other protected
sites under the Interagency Monitoring of Protected Visual
Environments (IMPROVE) net-work. Values are presented
Exhibit 2-26. Visibility in selected National Parks and Wilderness Areas in the U.S., 1992-2004ab
A. Western U.S. B. Eastern U.S.
300
250
200
150
100
50
Best visibility days____
Mid-range visibility days
Worst visibility days
* win!
kili-h/
Mid-range visibility days
Worst visibility days
'92 '93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '92 '93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04
Year
Monitoring sites
Coverage: 28 monitoring sites in the western U.S. and 10 monitoring sites in the eastern U.S. with sufficient
data to assess visibility trends from 1992 to 2004. V *..^*
, *- -
Visual ranges are calculated from the measured levels of different components within airborne particles and
these components' light extinction efficiencies. es
Data source: IMPROVE, 2007
30
-------
INDICATI
Regional Haze
for 10 Eastern (east of 100 degrees west longitude) sites
and 28 Western (west of 100 degrees west longitude) sites.
Visibility, expressed as visual range, is calculated from the
measured levels of different components within airborne
particles and these components' light extinction efficien-
cies. The IMPROVE algorithm (Debell et al, 2006)
includes an adjustment for ammonium sulfate and ammo-
nium nitrate to account for their adsorption of water vapor
from the atmosphere under elevated relative humidity
conditions. The IMPROVE particle data are generated
by laboratory analysis of 24-hour duration filter samples
collected at each site on a one-day-in-three schedule. This
indicator tracks visibility in three categories: worst visibil-
ity conditions (the average of the 20 percent worst visibility
days); best visibility conditions (the average of the 20 per-
cent best visibility days); and mid-range visibility condi-
tions (the average of the remaining 60 percent of days).
What the Data Show
On average, the best visibility in selected National Parks
and Wilderness Areas in the East, as calculated from the
measured concentrations of components of PM, is only
slightly better than the worst visibility in selected National
Parks and Wilderness Areas in the West (Exhibit 2-26).
In 2004, the average visual range for the worst days in the
East was 31 km (19 miles), compared to 137 km (85 miles)
for the best visibility days. In the West, the average visual
range in 2004 extended from 109 km (68 miles) on the
•worst days to 260 km (162 miles) on the best days. In both
regions, the average visual range in selected National Parks
and Wilderness Areas increased since 1992 for worst, mid-
range, and best visibility days. The increased visual ranges
between 1992 and 2004 for mid-range visibility days were
46 percent in the East and 14 percent in the West.
Indicator Limitations
• These data represent visibility in a sampling of selected
National Parks and Wilderness Areas and are not repre-
sentative of other rural or urban areas.
Data Sources
Summary data in this indicator were provided by the
National Park Service Air Quality Division, based on ambi-
ent air monitoring data collected as part of the IMPROVE
network (IMPROVE, 2007) and a computational algorithm
last updated in August 2007 (http://vista.cira.colostate.
edu/views/Web/IMPROVE/SummaryData.aspx). Vis-
ibility trends in this indicator are derived from the subset
of IMPROVE monitoring stations outside urban areas that
have sufficient data to assess trends between 1992 and 2004.
References
Debell, L.J., K.A. Gebhart, W.C. Malm, M.L. Pitchford,
B.A. Schichtel, and W.H. White. 2006. Spatial and seasonal
patterns and temporal visibility of haze and its constituents
in the United States: Report IV
IMPROVE (Interagency Monitoring of Protected Visual
Environments). 2007. Data from the IMPROVE network
based on the "New IMPROVE algorithm" (updated
August, 2007). Accessed 2007.
U.S. EPA (United States Environmental Protection
Agency). 2004a. The particle pollution report: Current
understanding of air quality and emissions through 2003.
EPA/454/R-04/002. Research Triangle Park, NC.
U.S. EPA. 2004b. The ozone report: Measuring progress
through 2003. EPA/454/K-04/001. Research Triangle
Park, NC.
U.S. EPA. 2003. Latest findings on national air quality
—2002 status and trends. EPA/454/K-03/001. Research
Triangle Park, NC.
31
-------
INDICATOR
Sulfur Dioxide Emissions
Sulfur dioxide (SO2) belongs to the family of sulfur
oxide (SO ) gases. These gases are formed when fuel
containing sulfur (mainly coal and oil) is burned (e.g.,
for electricity generation) and during metal smelting and
other industrial processes. High concentrations of SO2 are
associated with multiple health and environmental effects
(U.S. EPA, 2003). The highest concentrations of SO2 have
been recorded in the vicinity of large industrial facilities.
Although relatively few people live in areas where SO
concentrations exceed the National Ambient Air Qual-
ity Standards (NAAQS), SO2 emissions are an important
environmental issue because they are a major precursor to
ambient PM2 5 concentrations: many more people live in
PM2 5 non-attainment areas, which has several documented
human health and ecological effects (the PM Concentra-
tions indicator).
Health effects associated with SO depend on the expo-
sure concentrations and durations, and on the susceptibility
of exposed populations. Asthmatics are much more suscep-
tible to SO2 exposure than people who do not have asthma
(U.S. EPA, 1986). Effects associated with longer-term
exposures to high concentrations of SO , in conjunction
•with high levels of PM, include respiratory illness, altera-
tions in the lungs' defenses, and aggravation of existing
heart or lung disease. The most susceptible populations
under these conditions include individuals with cardiovas-
cular disease or chronic lung disease, children, and older
adults (U.S. EPA, 1982).
Many other environmental concerns are associated with
high concentrations of SO2. For example, airborne SO2,
along with NO , contributes to acidic deposition (the Acid
Deposition indicator); SO2 is a major precursor to PM2 5
(the PM Concentrations indicator); and SO2 contributes
to impaired visibility (the Regional Haze indicator). SO2
exposure also can harm vegetation by increasing foliar
injury, decreasing plant growth and yield, and decreasing
the number and variety of plant species in a given commu-
nity. Finally, SO can accelerate the corrosion of materials
(e.g., concrete, limestone) that are used in buildings, stat-
ues, and monuments that are part of the nation's cultural
heritage (U.S. EPA, 1982).
This indicator presents SO emissions from tradition-
ally inventoried anthropogenic source categories: (1) "Fuel
combustion: selected power generators," which includes
emissions from coal-, gas-, and oil-fired power plants that
are required to use continuous emissions monitors (CEMs)
to report emissions as part of the Acid Rain Program
(ARP); (2) "Fuel combustion: other sources," which
includes industrial, commercial, and institutional sources,
as well as residential heaters and boilers not required to use
CEMs; (3) "Other industrial processes," which includes
chemical production and petroleum refining; (4) "On-
road vehicles," which includes cars, trucks, buses, and
Exhibit 2-27. SC>2 emissions in the U.S. by
source category, 1990 and 1996-20023
On-road and
nonroad
mobile
sources
Fuel combustion:
selected power generators1
Fuel combustion: other sources
Year
"Data are presented for 1990 and 1996-2002, as datasets from
these inventory years are fully up to date. Data are available for
inventory years 1991-1995, but these data have not been updated
to allow comparison with data from 1990 and 1996-2002.
bThis category includes emissions from only those power plants
required to use continuous emissions monitors under the Acid
Rain Program.
Data source: U.S. EPA, 2007b
motorcycles; (5) "Nonroad vehicles and engines," which
include farm and construction equipment, lawnmowers,
chainsaws, boats, ships, snowmobiles, aircraft, and others.
Because a substantial portion of airborne SO comes from
fossil fuel combustion in electric utilities, this indicator
includes the separate "Fuel combustion: selected power
generators" category in addition to the four categories
presented in the other emissions indicators.
SO2 emissions data are tracked by the National Emis-
sions Inventory (NEI). The NEI is a composite of data
from many different sources, including industry and
numerous state, tribal, and local agencies. Different data
sources use different data collection methods, and many of
the emissions data are based on estimates rather than actual
measurements. For major electricity generating units, most
data come from CEMs that measure actual emissions. For
other fuel combustion sources and industrial processes,
data are estimated using emission factors. Emissions from
on-road and nonroad sources were estimated using EPA-
approved modeling approaches (U.S. EPA, 2007a).
NEI data have been collected since 1990 and cover all
50 states and their counties, D.C., the U.S. territories of
Puerto Rico and Virgin Islands, and some of the territories
of federally recognized American Indian nations. Data are
presented only for 1990 and from 1996 to 2002; prior to
1996, only the 1990 data have been updated to be compa-
rable to the more recent inventories.
32
-------
INDICATOR
Sulfur Dioxide Emissions
What the Data Show
National estimated SO2 emissions decreased 37 percent
between 1990 and 2002 (from 23,064,000 to 14,639,000
tons) (Exhibit 2-27). This downward trend resulted
primarily from emissions reductions at electric utilities.
Between 1990 and 2002, air emissions from electric utili-
ties have consistently accounted for roughly two-thirds of
the nation-wide SO2 emissions.
Net SO2 emissions declined in all EPA Regions between
1990 and 2002 (Exhibit 2-28). During this time frame, the
largest percent reductions in SO2 emissions were seen in
Regions 1 (59 percent), 2 (49 percent), and 5 (48 percent),
and the smallest reductions were observed in Regions 6 (15
percent) and 9 (18 percent).
Indicator Limitations
• Though emissions from most electric utilities are mea-
sured directly using continuous monitoring devices,
SO emissions data for other source types are based on
estimates that employ emission factors generated from
empirical and engineering studies. Although these esti-
mates are generated using well-established approaches,
the estimates have uncertainties inherent in the emission
factors and emissions models used to represent sources for
•which emissions have not been directly measured.
• Comparable SO2 emissions estimates through the NEI
are available only for 1990 and 1996-2002. Data for
1991-1995 are not provided due to differences in emis-
sions estimation methodologies from other inventory
years, which could lead to improper trend assessments.
• SO emissions from "miscellaneous sources" are not
included in the total emissions. Details on emissions
from miscellaneous sources can be found by download-
ing 2002 NEI inventory data for the "nonpoint sector"
(http: //www. epa.gov/ttn/chief/net/2002inventory.html).
• The methodology for estimating emissions is continually
reviewed and is subject to revision. Trend data prior to
these revisions must be considered in the context of
those changes.
• Not all states and local agencies provide the same data or
level of detail for a given year.
Data Sources
Summary data in this indicator were provided by EPA's
Office of Air Quality Planning and Standards, based on
SO2 emissions data in the NEI (U.S. EPA, 2007b)
(http://www.epa.gov/ttn/chief/net/2002inventory.html).
This indicator aggregates the NEI data by source category
and EPA Region.
Exhibit 2-28. SO2 emissions in the U.S. by EPA
Region, 1990 and 1996-20023
'96 '97 '98 '99 '00 '01 '02
Year
aData are presented for 1990
and 1996-2002, as datasets
from these inventory years are
fully up to date. Data are
available for inventory years
1991-1995, but these data have
not been updated to allow
comparison with data from
1990 and 1996-2002.
Data source: U.S. EPA, 2007b
EPA Regions
References
U.S. EPA (United States Environmental Protection
Agency). 2007a. Documentation for the final 2002 mobile
National Emissions Inventory, Version 3.
U.S. EPA. 2007b. Data from the 2002 National Emissions
Inventory, Version 3.0. Accessed 2007.
U.S. EPA. 2003. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
U.S. EPA. 1986. Second addendum to the air quality
criteria for particulate matter and sulfur oxides (1982):
Assessment of newly available health effects information.
EPA/450/S-86/012. Research Triangle Park, NC.
U.S. EPA. 1982. Air quality criteria for particulate mat-
ter and sulfur oxides. EPA/600/P-82/020a-c. Research
Triangle Park, NC.
33
-------
INDICATO
Acid Deposition
Every year, millions of tons of sulfur dioxide and nitro-
gen oxides are emitted to the atmosphere as a result of
the burning of fossil fuels and from other high tempera-
ture sources (the Sulfur Dioxide Emissions indicator; the
Nitrogen Oxides Emissions indicator). These gases react
with water, oxygen, and oxidants to form acidic com-
pounds, which may be carried hundreds of miles by the
•wind—even across state or national borders. Acid deposi-
tion occurs when these compounds fall to the Earth in one
of two forms: wet (dissolved in rain, snow, and fog) or dry
(solid and gaseous particles deposited on surfaces during
periods of no precipitation). While wet deposition is the
more widely recognized form (more commonly referred
to as "acid rain"), dry deposition can account for 20 to 80
percent of total acid deposition depending on location and
climate (MACTEC Engineering and Consulting, Inc.,
2005). In the environment, acid deposition causes soils
and water bodies to acidify, which can make the water
unsuitable for some fish and other wildlife. Some types of
ecosystems, those with less "buffering" capacity, are more
sensitive to acid deposition than others.
Scientists often use acid neutralizing capacity, a measure
of the amount of anions, protons, and non-proton cations
in the water, as an indicator of which lakes and streams are
most sensitive to acidification (NAPAP, 1991). Most surface
•waters in the West do not exhibit many symptoms of acidi-
fication, because relatively small amounts of acid deposition
occur in acid-sensitive regions. In the Northeast and along
the Appalachian Mountains, however, relatively high levels
of acid deposition occur in acid-sensitive regions, or regions
•without enough geochemical buffering capacity to prevent
acidification of surface waters by acid deposition (the Lake
and Stream Acidity indicator). Therefore, reductions in acid
deposition have the largest impact on acidification of lakes
and streams in those areas.
Acid deposition damages some trees, particularly at
high elevations, and speeds the decay of buildings, statues,
and sculptures that are part of our national heritage (U.S.
EPA, 2003). The nitrogen portion of acid deposition also
contributes to eutrophication in coastal ecosystems, the
symptoms of which include potentially toxic algal blooms,
fish kills, and loss of plant and animal diversity. Acidi-
fication of lakes and streams can increase the amount of
methylmercury available in aquatic systems (Winfrey and
Rudd, 1990). Finally, increased levels of sulfate in ground-
level air, a phenomenon related to dry deposition, can con-
tribute to decreased visibility as well as a variety of human
health problems (U.S. EPA, 2003).
Total acid deposition in this indicator is determined using
wet deposition measurements and dry deposition calculated
from ambient air concentration measurements. Wet depo-
sition is measured through chemical analysis of rainwater
collected at sites across the U.S. The primary source of wet
Exhibit 2-29. Wet sulfate (SO42~) deposition in
the contiguous U.S., 1989-1991 and 2004-20063
A. Average wet S042 deposition, 1989-1991
B. Average wet S042 deposition, 2004-2006
"Coverage: 169
monitoring sites in
1989-1991 and 202
monitoring sites in
2004-2006.
Data source: NADP,
2007
Wet S042 deposition
(kilograms per hectare):
12 16 20 24 28>32
• Monitoring site
deposition information comes from the National Atmo-
spheric Deposition Program/National Trends Net-work.
The chemical components of wet deposition include sulfate,
nitrate, and ammonium. Dry deposition is not measured
directly. EPA's Clean Air Status and Trends Net-work deter-
mines dry deposition inferentially by measuring ambient
air concentrations of acidic compounds and then calculat-
ing deposition rates using a multi-layer model that depends
on meteorological data collected at the sites as -well as local
vegetative conditions (http://www.epa.gov/castnet/).
Chemicals measured include components of particulate
matter (sulfate [SO42~] and nitrate [NO3~]), gaseous nitric
acid (HNO3), sulfur dioxide (SO2), ammonia (NH3), and
ammonium (NH4+).
34
-------
INDICATI
Acid Deposition
This indicator uses the 3-year average from 1989-1991
as a baseline, as this period immediately predates con-
trols on sulfur and nitrogen oxide emissions mandated by
the 1990 Clean Air Act Amendments. Baseline data are
compared to the most recent 3-year average data available
(2004-2006). Use of 3-year average data helps ensure that
trends reflect actual changes in acid deposition, instead of
shorter-term fluctuations in meteorological conditions.
Additionally, this indicator presents annual trend data for
total deposition, which characterizes deposition over the
entire period of record, not just for the baseline and most
recent 3-year average periods.
What the Data Show
Wet Deposition Trends
Analyses of long-term monitoring data from the National
Atmospheric Deposition Program show that wet deposition
of both sulfur and nitrogen compounds has decreased over
the last 17 years (Exhibits 2-29 and 2-30).
Wet sulfate deposition decreased across much of the U.S.
during the 1990s (Exhibit 2-29). The greatest reductions
in wet sulfate deposition occurred in the Mid-Appalachian
region (Maryland, New York, West Virginia, Virginia,
and most of Pennsylvania) and the Ohio River Valley. Less
dramatic reductions were observed across much of New
England and portions of the Southern Appalachians. Aver-
age regional decreases in wet deposition of sulfate between
the periods 1989-1991 (panel A) and 2004-2006 (panel B)
•were approximately 35 percent in the Northeast, 33 percent
in the Midwest, 28 percent in the Mid-Atlantic, and 20
percent in the Southeast.
Wet nitrate deposition decreased approximately 33 per-
cent across the Northeast and 27 percent in the Mid-Atlan-
tic between the periods 1989-1991 (Exhibit 2-30, panel A)
and 2004-2006 (panel B). However, there is a high degree
of variability in the measurements used to calculate these
percentages, complicating efforts to reliably estimate trends
for wet nitrate deposition. Wet deposition of inorganic
nitrogen has not changed substantially in the rest of the
country over this period.
Total Deposition Trends
As with wet deposition, total deposition (the sum of wet
and dry deposition) decreased between 1989-1991 and
2004-2006, and reductions were more substantial for
sulfur compounds than for nitrogen compounds (Exhibits
2-31 and 2-32). In the eastern U.S., where data are most
abundant, total sulfur deposition decreased by 36 percent
between 1990 and 2005 (Exhibit 2-33), while total nitro-
gen deposition decreased by 19 percent over the same time
frame (Exhibit 2-34). Note that total nitrogen deposition
in this indicator does not include nitrogen components,
such as ammonia, which can be a significant portion of the
dry deposition.
Exhibit 2-30. Wet nitrate (NOs) deposition in
the contiguous U.S., 1989-1991 and 2004-20063
A. Average wet N03 deposition, 1989-1991
B. Average wet N03 deposition, 2004-2006
•
"Coverage: 169
monitoring sites in
1989-1991 and 202
monitoring sites in
2004-2006.
Data source: NADP,
2007
Wet NOs deposition
(kilograms per hectare):
02468 10121416182022>24
•Monitoring site
Indicator Limitations
• Geographic coverage is limited, particularly for dry depo-
sition (and thus total deposition as well), but the concentra-
tion of sites in the Midwest and Northeast is justified by
the fact that acid rain is much more of a problem in those
regions than it is in the West, Great Plains, or Southeast.
• Measurement techniques for dry deposition have improved
substantially, but characterization of dry deposition still
requires a combination of measurements and modeling,
•which has inherent uncertainties. Further, dry deposition
presented in this indicator does not include contributions
from deposition of gaseous ammonia.
35
-------
INDICATO
Acid Deposition
Exhibit 2-31. Total sulfur deposition in the contiguous U.S., 1989-1991 and 2004-20063
A. Average total sulfur deposition, 1989-1991
B. Average total sulfur deposition, 2004-2006
\, ir
-------
INDICATO
Acid Deposition
Exhibit 2-32. Total nitrogen deposition in the contiguous U.S., 1989-1991 and 2004-20063
A. Average total nitrogen deposition, 1989-1991
Numbers indicate total nitrogen deposition (kilograms per hectare), averaged
1(. over a 3-year period.
f =g-10 Sizes of circles indicate the relative magnitude of total nitrogen deposition.
Colors in circles indicate the breakdown of total nitrogen deposition:
• Dry nitrogen deposition • Wet nitrogen deposition
B. Average total nitrogen deposition, 2004-2006
Numbers indicate total nitrogen deposition (kilograms per hectare), averaged
over a 3-year period.
_15 Sizes of circles indicate the relative magnitude of total nitrogen deposition.
~5 Colors in circles indicate the breakdown of total nitrogen deposition:
Dry HN03deposition Dry N03~ deposition Wet N03~ deposition
Dry NIV deposition "Wet NIV deposition
3.6
"Coverage: 37 monitoring sites in 1989-1991 and 73 monitoring sites in 2004-2006.
Data source: NADP, 2007; U.S. EPA, 2007
37
-------
INDICATO
Acid Deposition
Exhibit 2-33. Total sulfur deposition in the
eastern United States, 1990-20053
25
<= 5
11 2°
t 22" „
i_ _cs To
•^ s 10
i»
« E
12 S.
0
90% of sites have annual sulfur deposition
below this line
10% of sites have annual
sulflur deposition below this line
'90 '91 '92 '93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '05
Year
Coverage: 34 monitoring sites in the eastern United States.
Data source: MACTEC Engineering and Consulting, inc., 2006
Data Sources
Summary data in this indicator were provided by EPA's
Office of Atmospheric Programs, based on deposition
data from two sources. Wet deposition data are from the
National Atmospheric Deposition Program/National Trends
Network (NADP, 2007) (http://nadp.sws.umc.edu/), and
dry deposition data are from the Clean Air Status and
Trends Network (U.S. EPA, 2007) (http://www.epa.gov/
castnet). This indicator aggregates data across 3-year periods
to avoid influences from short-term fluctuations in meteoro-
logical conditions, and wet deposition data were interpolated
among monitoring stations to generate the maps shown in
Exhibits 2-29 and 2-30.
References
MACTEC Engineering and Consulting, Inc. 2006. Clean
Air Status and Trends Network (CASTNET): 2005 annual
report. Prepared for U.S. EPA, Office of Air and Radiation.
MACTEC Engineering and Consulting, Inc. 2005. Clean
Air Status and Trends Network (CASTNET): 2004 annual
report. Prepared for U.S. EPA, Office of Air and Radiation.
Exhibit 2-34. Total nitrogen deposition in the
eastern United States, 1990-2005a
12
90% of sitesjiave annual nitrogen deposition
below this line
If 2
10% of sites have annual nitrogen deposition
below this line
'90 '91 '92 '93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '05
Year
Coverage: 34 monitoring sites in the eastern United States.
Data source: MACTEC Engineering and Consulting, Inc., 2006
NADP (National Atmospheric Deposition Program). 2007.
Data from the NADP/National Trends Network.
Accessed 2007.
NAPAP (National Acid Precipitation Assessment Program).
1991. 1990 integrated assessment report. Washington, DC.
U.S. EPA (United States Environmental Protection
Agency). 2007. Data from the Clean Air Status and Trends
Network. Accessed 2007.
U.S. EPA. 2003. Latest findings on national air quality:
2002 status and trends. EPA/454/K-03/001. Research
Triangle Park, NC.
Winfrey, M.R., andJW.M. Rudd. 1990. Environmental
factors affecting the formation of methyl mercury in low
pH lakes. Environ. Toxicol. Chem. 9(7):853-869.
38
-------
Percent of Days with Air Quality Index Values
Greater Than 100
The Air Quality Index (AQI) provides information
on pollutant concentrations of ground-level ozone,
particulate matter, carbon monoxide, sulfur dioxide, and
nitrogen dioxide. Formerly known as the Pollutant Stan-
dard Index, the nationally uniform AQI is used by state
and local agencies for reporting daily air quality and air
quality related health advisories to the public.
In 1999, the AQI was updated to reflect the latest sci-
ence on air pollution health effects and to make it more
appropriate for use in contemporary news media (U.S.
EPA, 2003a). It also serves as a basis for community-based
programs that encourage the public to take action to
reduce air pollution on days when levels are projected to
be of concern. The index has been adopted by many other
countries (e.g., Mexico, Singapore, Taiwan) to provide the
public with information on air quality.
The AQI is based on pollutant concentration data
measured by the State and Local Air Monitoring Stations
net-work and by other special purpose monitors. The AQI
is monitored in city groupings known as metropolitan
statistical areas (MSAs), which are defined by the Office of
Management and Budget. For most pollutants in the index,
the concentration is converted into index values between 0
and 500, "normalized" so that an index value of 100 repre-
sents the short-term, health-based standard for that pollut-
ant as established by EPA (U.S. EPA, 1999). The higher the
index value, the greater the level of air pollution and health
risk. An index value of 500 reflects a risk of imminent and
substantial endangerment of public health. The level of
the pollutant with the highest index value is reported as
the AQI level for that day. An AQI value greater than 100
means that at least one criteria pollutant has reached levels
at which people in sensitive groups may experience health
effects. A complete description of how AQI values are
calculated and what they represent is documented in many
publications (e.g., U.S. EPA, 2003b).
This indicator is based on the percent of days across 93
large MSAs (500,000 people or more) during the year that
recorded an AQI greater than 100 at one or more monitor-
ing sites in the MSA. While the AQI indicator is calculated
from ambient concentration data for criteria pollutants,
this indicator's trends should not be expected to mirror
the trends in the other ambient concentration indicators,
due to the differing spatial coverage of monitoring stations
across the various indicators.
The percent of days with AQI greater than 100 was cal-
culated in two steps. First, for each year, the total number
of days with AQI above 100 in each of the 93 MSAs was
summed in order to get a national total. Then, the national
total was divided by the total number of days in the annual
sample (365 X 93, or 33,945 days) to obtain the percent-
age of days with AQI above 100 in a year. Note that this
Exhibit 2-37. Percent of days with Air Quality
Index (AQI) greater than 100 in selected U.S.
metropolitan areas, 1990-2006ab
A. AQI trend based on all criteria pollutants (1990-2006)°
1999 was the first year PM2.5
was included in the AQI 2004-2006:3-year
1990-1992: 3-year average = 2.8%
average = 4.5%
B. AQI trend based on ozone (1990-2006)
1990-1992: 3-year
average = 4.3%
2004-2006: 3-year
average = 1.9%
00 '02
'04
'06
Year
Coverage: 93 metropolitan
areas for AQI trend based on
all criteria pollutants, 90
metropolitan areas for AQI
trend based on ozone, and 89
metropolitan areas for AQI
trend based on PM2.5.
bFor each MSA, the percentage
of days with AQI greater than
100 was calculated by dividing
the number of days per year
with AQI greater than 100 by
365 total days. However,
because PM2.5 is not
monitored daily in some areas,
the actual percentage of days
with AQI greater than 100
C. AQI trend based
on PM2.5 (1999-2006)"
'i 5
re 4
2004-2006: 3-year
average = 1.1%
1999-2001: 3-year
average = 2.1%
'00
'02 '04
Year
'06
might be higher than what is shown in Panels A and C.
=l_ead does not factor into the AQI calculation for all criteria pollutants.
dData for 1990-1998 are not shown because 1999 was the first year
that PM2.5 was included in the AQI.
Data source: U.S. EPA, 2007
39
-------
INDICATO
Percent of Days with Air Quality Index Values
Greater Than 100
calculation will understate the actual percentage of days
•with AQI above 100 for pollutants that are not measured
daily (e.g., PM25).
Data are presented for 1990 through 2006. However,
because meteorology can strongly influence AQI values in
a given year, the change in AQI over time is evaluated by
comparing the 3-year average observation at the begin-
ning of the period of record (i.e., 1990-1992) to the 3-year
average at the end (i.e., 2004-2006). Comparing 3-year
averages reduces the potential for biases introduced by
years with unique meteorological conditions. The air qual-
ity data that go into the index consist of daily (24-hour)
measurements for PM1Q and PM2 5 and continuous (1-hour)
measurements for CO, NO2, ozone, and SO2. Lead mea-
surements do not factor into the AQI. Of the pollutants
considered, only four (CO, ozone, PM, and SO2) usually
exhibit AQI values greater than 100.
What the Data Show
AQI Based on All Criteria Pollutants (Except Lead)
The percent of days with AQI greater than 100 in 93
large MSAs based on all criteria pollutants (except lead)
decreased from 4.5 over the 1990-1992 time frame to
2.8 over the 2004-2006 time frame (Exhibit 2-37, panel
A). The AQI data based on all criteria pollutants are not
directly comparable over this time frame, because PM2 5
measurements started to factor into the index in 1999. For
this reason, the indicator also presents AQI trends based
strictly on ozone and PM2 measurements.
AQI Based on Ozone Only
For a nearly identical subset of MSAs, the percent of days
•with AQI values greater than 100 due to ozone levels alone
(based on the 1997 NAAQS) decreased from 4.3 over the
1990-1992 time frame to 1.9 over the 2004-2006 time
frame (Exhibit 2-37, panel B). Before PM2 5 became part of
the index in 1999, ozone typically accounted for more than
90 percent of the days with AQI greater than 100.
AQI Based on PM25 Only
In the 1999-2001 period, PM2 concentrations accounted
for 2.1 percent of days with AQI greater than 100. This
contribution decreased in subsequent years, falling to 1.1
percent for the 2004-2006 period.
AQI in the EPA Regions Based on All Criteria Pollutants
(Except Lead)
Trends in AQI based on all criteria pollutants (except lead)
between 1990 and 2006 varied across the ten EPA Regions
(Exhibit 2-38). For nine of the Regions, the percent of
days \vith AQI greater than 100 in 2006 was lower than
that in 1990, though substantial year-to-year variability
Exhibit 2-38. Percent of days with Air Quality
Index (AQI) greater than 100 in selected U.S.
metropolitan areas by EPA Region, 1990-2006ab
'91 '92 '93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '05 '06
Year
Coverage: 93 metropolitan areas.
bTrend is based on AQI data for
all criteria pollutants, except for
lead. Note that 1999 was the
first year that PM2.5 was
included in the AQI.
Data source: U.S. EPA, 2007
EPA Regions
occurred. In Region 8, the percent of days with AQI
greater than 100 in 2006 was higher than that observed in
1990. However, as noted above, the AQI values for 1990
and 2006 are not directly comparable, because PM2 B mea-
surements did not factor into AQI prior to 1999.
Indicator Limitations
• The AQI does not address hazardous air pollutants.
• Air quality can vary across a single MSA. In assigning a
single number for each pollutant in each MSA, the AQI
does not reflect this potential variation.
• The data for this indicator are limited to MSAs compris-
ing urban and suburban areas with populations greater
than 500,000. Thus, this indicator does not reflect MSAs
smaller than 500,000 or rural areas.
• The AQI does not show which pollutants are causing
the days with an AQI of more than 100, or distinguish
between days with AQI slightly above 100 and days with
much higher AQI.
• This composite AQI indicator does not show which
specific MSAs, or how many MSAs, have problems—a
specific number of days could reflect a few areas with per-
sistent problems or many areas with occasional problems.
40
-------
Percent of Days with Air Quality Index Values
Greater Than 100
• This indicator only covers the days on which ambi-
ent monitoring occurred. Because PM2 5 is not sampled
daily in some areas, the data presented in this indicator
may understate the actual number of days on which AQI
values were greater than 100 due to PM2 5 concentrations.
Although ozone is not sampled throughout the year, the
percent of days with AQI greater than 100 is believed to
be accurate because monitoring occurs throughout the
summer, when ozone concentrations are highest.
Data Sources
Summary data in this indicator were provided by EPA's
Office of Air Quality Planning and Standards, based on
AQI values computed from ambient air monitoring data for
criteria pollutants found in EPA's Air Quality System (U.S.
EPA, 2007). Spreadsheets with the processed AQI data for
the 93 MSAs considered in this indicator are publicly avail-
able (http://www. epa.gov/air/airtrends/factbook.html).
This indicator aggregates the processed AQI data nation-
ally and by EPA Region.
References
U.S. EPA (United States Environmental Protection Agency).
2007. Data from the Air Quality System. Accessed 2007.
U.S. EPA. 2003a. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
U.S. EPA. 2003b. Air Quality Index: A guide to air quality
and your health. EPA-454/K-03-002.
U.S. EPA. 1999. Air quality index reporting, 40 CFRpart 58.
41
-------
Air Toxics Emissions
Toxic air pollutants, also known as air toxics or hazard-
ous air pollutants (HAPs), are those pollutants that
are known or suspected to cause cancer or are associated
•with other serious health (e.g., reproductive problems,
birth defects) or ecological effects. Examples of air tox-
ics include benzene, found in gasoline; perchloroethylene,
emitted from some dry cleaning facilities; and methylene
chloride, used as a solvent by a number of industries. Most
air toxics originate from anthropogenic sources, including
mobile sources (e.g., cars, trucks, construction equipment),
stationary sources (e.g., factories, refineries, power plants),
and indoor sources (e.g., building materials, cleaning
solvents). Some air toxics are also released from natural
sources such as volcanic eruptions and forest fires. Second-
ary formation of certain air toxics, such as acetaldehyde
and formaldehyde, can also occur when precursor chemi-
cals react in the atmosphere. The Clean Air Act identifies
188 air toxics associated with industrial sources. Twenty
of these air toxics also are associated with mobile sources
(U.S. EPA, 2003).
People \vho inhale certain air toxics at sufficient con-
centrations may experience various health effects, including
cancer, damage to the immune system, and neurological,
reproductive (e.g., reduced fertility), developmental, or
respiratory health problems (CDC, 2005). Air toxics also
can present risks through other exposure path-ways. For
example, air toxics may deposit onto soils or surface -waters,
•where they can then enter the food web and may eventu-
ally be ingested by humans. Plants and animals also may be
harmed by exposures to air toxics (U.S. EPA, 2003).
Air toxics emissions data are tracked by the National
Emissions Inventory (NEI). The NEI is a composite of
data from many different sources, including industry and
numerous state, tribal, and local agencies. Different data
sources use different data collection methods, and many of
the emissions data are based on estimates rather than actual
measurements. For most fuel combustion sources and
industrial sources, emissions are estimated using emission
factors. Emissions from on-road and nonroad sources were
estimated using EPA-approved modeling approaches (U.S.
EPA, 2007a).
NEI data have been collected since 1990 and cover all 50
states and their counties, D.C., the U.S. territories of Puerto
Rico and the Virgin Islands, and some of the territories of
federally recognized American Indian nations. The NEI
includes baseline air toxics data for the 1990-1993 period
and since then has been updated every 3 years. The baseline
period represents a mix of years depending on data availabil-
ity for various source types. While NEI data for air toxics
•were also compiled for 1996 and 1999, the methodology
used in those years for air toxics differed considerably from
the methodology that was used in 2002. Therefore, the 1996
and 1999 data are not presented because comparing the two
inventories might lead to invalid conclusions.
Exhibit 2-40. Air toxics emissions in the U.S. by
source category, 1990-1993 and 2002
1— 7
CD
CD 6
Q.
| 4
I 3
I 2
'i 1
LLJ
0
n Fires (prescribed
burns and
wildfires)
n Nonroad vehicles
and engines
E On-road vehicles
n Stationary
sources (not
including fires)
1990-1993d
2002
Year
a1990-1993 is considered the baseline period for air toxics
emissions. The baseline period spans multiple years due to the
availability of emissions data for various source categories. The
data presented for the baseline period are annual emissions (tons
per year) and are therefore comparable to the 2002 data.
Data source: U.S. EPA, 2007b
This indicator first presents emissions data for all air
toxics combined, both at the national level and broken
down into the ten EPA Regions. Consistent with the other
emissions indicators, the national data are organized into
the following source categories: (1) "Stationary sources,"
•which include fuel combustion sources (coal-, gas-, and
oil-fired power plants; industrial, commercial, and institu-
tional sources; as well as residential heaters and boilers) and
industrial processes (chemical production, petroleum refin-
ing, and metals production) categories; (2) "Fires: prescribed
burns and wildfires," for insights on contributions from
some natural sources; (3) "On-road vehicles," which include
cars, trucks, buses, and motorcycles; and (4) "Nonroad
vehicles and engines," such as farm and construction equip-
ment, lawnmowers, chainsaws, boats, ships, snowmobiles,
aircraft, and others.
In addition to presenting emissions data aggregated
across all 188 air toxics, the indicator presents emissions
trends for five individual air toxics: acrolein, benzene,
1,3-butadiene, ethylene dibromide, and hydrazine. These
compounds were selected for display because EPA's 1999
National Air Toxics Assessment estimates that they pres-
ent the greatest nation-wide health risks (whether for cancer
or non-cancer endpoints) among the subset of air toxics
for \vhich available emissions and toxicity data supported
an evaluation (U.S. EPA, 2006). This indicator breaks the
emissions data for these five air toxics into multiple source
categories, -with the most appropriate categories for display
purposes differing from one air toxic to the next.
42
-------
Air Toxics Emissions
What the Data Show
Trends Aggregated Across All 188 Air Toxics
According to NEI data, estimated annual emissions for the
188 air toxics combined decreased 36 percent, from 7.2
million tons per year in the baseline period (1990-1993)
to 4.6 million tons per year in 2002 (Exhibit 2-40). This
downward trend resulted primarily from reduced emissions
from stationary sources and on-road mobile sources.
In 2002, air toxics emissions in the ten EPA Regions
ranged from 166,000 tons in Region 1 to 1,056,000 tons
in Region 4 (Exhibit 2-41). Regional trends cannot be
characterized, because a complete set of state and local air
toxics emissions data are not available for the 1990-1993
baseline period.
Trends for Selected Air Toxics
Exhibit 2-42 shows emissions trends for five compounds
believed to account for the greatest health risks that are
attributed to air toxics, according to a recent modeling
study (U.S. EPA, 2006). The five plots in this exhibit show
how emissions trends vary from compound to compound.
Estimated emissions decreased between the baseline period
(1990-1993) and 2002 for all five selected air toxics: acrolem
(51 percent decrease; see panel A), benzene (17 percent; panel
B), 1,3-butadiene (38 percent; panel C), ethylene dibromide
(63 percent; panel D), and hydrazine (84 percent; panel E).
Indicator Limitations
• The emissions data are largely based on estimates.
Although these estimates are generated using -well-
established approaches, the estimates have inherent
uncertainties. The methodology for estimating emissions
is continually reviewed and is subject to revision. Trend
data prior to any revisions must be considered in the
context of those changes.
• The indicator is an aggregate number that represents
contributions from 188 different chemicals with widely
varying toxicities and human exposures. Therefore,
the nation-wide trend for total air toxics and the result-
ing health effects likely differs from emissions trends for
specific chemicals. Similarly, because the indicator is a
nation-wide aggregate statistic, the trend may not reflect
emissions trends for specific locations.
• Not all states and local agencies provide the same data or
level of detail for a given year.
• There is uncertainty associated -with identifying -which
air toxics account for the greatest health risk nation-wide.
Exhibit 2-41. Air toxics emissions in the U.S. by
EPA Region, 2002
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10
EPA Region
Data source: U.S. EPA, 2007b
EPA Regions
Toxicity information is not available for every compound,
and emissions and exposure estimates used to character-
ize risk have inherent uncertainties. Additional limitations
associated -with the National Air Toxics Assessment are
well documented (U.S. EPA, 2006).
Data Sources
Summary data in this indicator -were provided by EPA's
Office of Air Quality Planning and Standards, based on
air toxics emissions data in the NEI (U.S. EPA, 2007b)
(http://www.epa.gov/ttn/chief/net/2002inventory.html).
This indicator aggregates the NEI data by source category,
EPA Region, and selected air toxics.
References
CDC (Centers for Disease Control and Prevention). 2005.
Third national report on human exposure to environ-
mental chemicals. NCEH Pub. No. 05-0570. Accessed
September 9, 2005.
43
-------
Air Toxics Emissions
Exhibit 2-42. Emissions of selected air toxics in the U.S. by source category, 1990-1993 and 2002a
-c- 70
fD
? 60
03
o.
22 50
_o
! 40
I 30
^ 20
1 10
E
LLJ g
-c- 90
ns
£ 80
a /o
O>
o 60
1 50
ns
I 40
£ 30
o>
| 20
'o>
.22 10
40
-c- 35
ns
£ 30
S 25
«
, 20
.22 10
E
m 5
A. Acrolein emissions
n Other sources
nNonroad vehicles
and engines
nOn-road vehicles
n Structural fires
n Prescribed burns
and wildfires
1990-1993 2002
Year
C. 1,3-Butadiene emissions
nNonroad vehicles
and engines
n Prescribed burns
and wildfires
1990-1993'
2002
Year
E. Hydrazine emissions
n Miscellaneous organic
chemical manufacturing
n Other sources
n Industrial organic
chemical production
• Industrial inorganic
chemical production
1990-1993
2002
Year
B. Benzene emissions
— 600
cs
U.S. EPA. 2007b. Data from the 2002 National Emissions
Inventory, Version 3.0. Accessed 2007.
U.S. EPA. 2006. 1999 national-scale air toxics assessment.
February.
U.S. EPA. 2003. National air quality and emissions trends
report—2003 special studies edition. EPA/454/R-03/005.
Research Triangle Park, NC.
44
-------
INDICATO
High and Low Stream Flows
Flow is a critical aspect of the physical structure of stream
ecosystems (Poff and Allan, 1995; Robinson et al.,
2002). High flows shape the stream channel and clear silt
and debris from the stream, and some fish species depend
on high flows for spawning. Low flows define the smallest
area available to stream biota during the year. In some cases,
the lowest flow is no flow at all—particularly in arid and
semi-arid regions where intermittent streams are common.
Riparian vegetation and aquatic life in intermittent streams
have evolved to complete their life histories during periods
•when water is available; however, extended periods of no
flow can still impact their survival (Fisher, 1995). Changes
in flow can be caused by dams, water withdrawals, ground
Exhibit 3-1. Changes in high flow in rivers and streams of
the contiguous U.S., 1961-2006, compared with 1941-1960
baseline3"
50
40
30
° 20
10
A. Increased high flow volume
>30% increase
— Non-reference
streams
— Reference
streams
>60% increase
—Non-reference
streams
— Reference
streams
1960
1970
1980 1990
Year
2000
2010
50
M40
&
'co
I 30
CD
to
2 20
d
CD
°- 10
0
B. Decreased high flow volume
>30% decrease
— Non-reference
streams
Reference
streams
>60% decrease
—Non-reference
streams
Reference
streams
1960
1970
1980 1990
Year
2000
2010
Coverage: 1,719 stream gauging sites (712 reference, 1,007 non-reference) in the
contiguous U.S. with flow data from 1941 to 2006. Reference streams have not
been substantially affected by dams and diversions; non-reference streams may or
may not have been affected in this way.
bBased on the annual 3-day high flow. For each stream site, the median high flow
was determined over a rolling 5-year window, then compared against the baseline.
Results are plotted at the midpoint of each window. For example, the value for
2002-2006 is plotted at the year 2004.
Data source: Heinz Center, 2007
•water pumping (which can alter base flow), changes in land
cover (e.g., deforestation or urbanization), and weather and
climate (Calow and Petts, 1992).
This indicator, developed by the Heinz Center (in
press), describes trends in stream flow volumes based on
daily flow data collected by the U.S. Geological Survey's
(USGS's) nation-wide net-work of stream flow gauging sites
from 1961 to 2006.
The first part of this indicator describes trends in high
flow volume, low flow volume, and variability of flow in
streams throughout the contiguous 48 states, relative to a
baseline period of 1941-1960. Data -were collected at two
sets of USGS stream gauging stations: a set of approxi-
mately 700 "reference" streams that have not
been substantially affected by dams and diver-
sions and have had little change in land use
over the measurement period, and a separate
set of approximately 1,000 "non-reference"
streams that reflect a variety of conditions (the
exact number of sites -with sufficient data var-
ies from one metric to another). The indicator
is based on each site's annual 3-day high flow
volume, 7-day low flow volume, and variabil-
ity (computed as the difference between the
1st and 99th percentile 1-day flow volumes in a
given year, divided by the median 1-day flow).
Annual values for each metric -were exam-
ined using a rolling 5-year window to reduce
the sensitivity to anomalous events. For each
site, the median value for the 5-year win-
dow -was compared to the median value for
the 1941-1960 baseline period. The indicator
shows the proportion of sites -where high flow,
low flow, or variability of flow -was more than
30 percent higher or 30 percent lower than the
baseline. It also shows differences of more than
60 percent.
This indicator also examines no-flow peri-
ods in streams in grassland and shrubland areas
of the contiguous 48 states. Data represent 280
USGS "reference" and "non-reference" stream
gauging sites in -watersheds -with at least 50
percent grass or shrub cover, as defined by the
2001 National Land Cover Database (NLCD)
(MRLC Consortium, 2007). The indicator
reports the percentage of these streams -with at
least one no-flow day in a given year, aver-
aged over a rolling 5-year window. Results are
displayed for all grassland/shrubland streams,
as -well as for three specific ecoregion divisions
(Bailey, 1995). This indicator also reports on
the duration of no-flow periods. For a subset
of 163 grassland/shrubland streams that had
at least one no-flow day during the study
45
-------
INDICATO
High and Low Stream Flows
period, the duration of the maximum no-
flow period in each year was averaged over
a rolling 5-year window and compared with
the average no-flow duration for the same site
during the 1941-1960 baseline period. A no-
flow period more than 14 days longer than
the baseline was described as a "substantial
increase"; a no-flow period more than 14 days
shorter than the baseline was classified as a
"substantial decrease."
What the Data Show
In an average year during the period of
record, roughly 20 percent of streams had
increases in high flow volume of more than
30 percent, relative to the 1941-1960 baseline
(Exhibit 3-1, panel A). A similar percent-
age had decreases of more than 30 percent
(Exhibit 3-1, panel B). Large fluctuations in
high flow volume are apparent over time,
•with both sets of trends suggesting relatively
wet periods in the early 1980s and mid-1990s
and relatively dry periods around 1990 and
the early 2000s. Reference and non-reference
stream sites show similar patterns, although
larger decreases in high flow volume were
more common in the non-reference streams.
Since the early 1960s, more streams
have shown increases in low flow volumes
than have shown decreases, relative to the
1941-1960 baseline period (Exhibit 3-2).
Among the many streams with larger low
flows are a few (2 to 4 percent in an aver-
age year) with increases of more than 600
percent. Fluctuations over time are apparent,
and while not as pronounced as the shifts in
high flow (Exhibit 3-1), they generally tend
to mirror the same relatively wet and dry
periods. Reference and non-reference streams
show similar low flow patterns over time, but
reference sites are less likely to have experi-
enced decreases in low flow.
Except for a few brief periods in the mid-1960s and again
around 1980, decreased flow variability has been much more
common than increased variability (Exhibit 3-3). Refer-
ence and non-reference streams have shown similar patterns
in variability over time, although reference streams were
slightly less likely to experience changes overall.
In areas with primarily grass or shrub cover, roughly
15 to 20 percent of stream sites typically have experienced
periods of no flow in a given year (Exhibit 3-4). Overall,
the number of streams experiencing no-flow periods has
declined slightly since the 1960s. Streams in the California/
Mediterranean ecoregion have shown the greatest decrease
Exhibit 3-2. Changes in low flow in rivers and streams of
the contiguous U.S., 1961-2006, compared with 1941-1960
baseline3"
A. Increased low flow volume
1970
1980
1990
2000
2010
Year
. Decreased low flow volume
>30% increase
— Non-reference
streams
— Reference
streams
>60% increase
— Non-reference
streams
— Reference
streams
>600% increase
— Non-reference
streams
— Reference
streams
>30% decrease
— Non-reference
streams
— Reference
streams
>60% decrease
— Non-reference
streams
Reference
streams
1970
1980 1990
Year
2000
2010
Coverage: 1,609 stream gauging sites (673 reference, 936 non-reference) in the
contiguous U.S. with flow data from 1941 to 2006. Reference streams have not
been substantially affected by dams and diversions; non-reference streams may or
may not have been affected in this way.
bBased on the annual 7-day low flow. For each stream site, the median low flow was
determined over a rolling 5-year window, then compared against the baseline.
Results are plotted at the midpoint of each window. For example, the value for
2002-2006 is plotted at the year 2004.
Data source: Heinz Center, 2007
in no-flow frequency, but they still experience more no-
flow periods than streams in the other two major grassland/
shrubland ecoregion divisions. Among grassland/shrubland
streams that have experienced at least one period of no flow
since 1941, more streams have shown a substantial decrease
in the duration of no-flow periods (relative to the 1941-1960
baseline) than a substantial increase (Exhibit 3-5).
Indicator Limitations
• The 1941-1960 baseline period was chosen to maxi-
mize the number of available reference sites and should
provide a sufficiently long window to account for natural
variability (Heinz Center, in press); however, it does not
46
-------
INDICATO
High and Low Stream Flows
50
40
30
° 20
10
1960
necessarily reflect "undisturbed" conditions.
Many darns and waterworks had already
been constructed by 1941, and other anthro-
pogenic changes (e.g., urbanization) were
already -widespread.
• Although the sites analyzed here are spread
•widely throughout the contiguous U.S.,
gauge placement by USGS is not a random
process. Gauges are generally placed on larger,
perennial streams and rivers, and changes seen
in these larger systems may differ from those
seen in smaller streams and rivers.
• This indicator does not characterize trends
in the timing of high and low stream flows,
•which can affect species migration, repro-
duction, and other ecological processes.
Data Sources
The data presented in this indicator were pro-
vided by the Heinz Center (2007), which con-
ducted this analysis for a forthcoming update
to its report, The State of the Nation's Ecosystems
(Heinz Center, in press). Underlying stream
flow measurements can be obtained from the
USGS National Water Information System
database (USGS, 2007) (http://waterdata.usgs.
gov/nwis).
References
Bailey, R.G. 1995. Description of the ecore-
gions of the United States. Second edition.
Misc. Publ. No. 1391 (rev). Washington, DC:
USDA Forest Service,
Calow, P., and G.E. Petts, eds. 1992.
The rivers handbook: Hydrological and
ecological principles. Volume 1. Oxford, UK:
Black-well Scientific.
Fisher, S.G. 1995. Stream ecosystems of the
•western United States. In: Gushing, C.E.,
KW. Cummings, and GW. Minshall, eds. River and
stream ecosystems, ecosystems of the world 22. New York,
NY: Elsevier.
Heinz Center (The H. John Heinz III Center for Science,
Economics, and the Environment). 2007. Data provided to
EPA by Anne Marsh, Heinz Center. October 15, 2007.
Heinz Center. In press. The state of the nation's ecosys-
tems: Measuring the lands, waters, and living resources of
the United States. 2007 update.
Exhibit 3-3. Changes in flow variability in rivers and streams
of the contiguous U.S., 1961-2006, compared with
1941-1960 baselineab
A. Increased flow variability
>30% increase
— Non-reference
streams
Reference
streams
>60% increase
—Non-reference
streams
Reference
streams
1970
1980
1990
2000
2010
50
40
30
° 20
10
Year
B. Decreased flow variability
>30% decrease
— Non-reference
streams
Reference
streams
>60% decrease
—Non-reference
streams
Reference
streams
1960
1970
1980 1990
Year
2000
2010
Coverage: 1,754 stream gauging sites (733 reference, 1,021 non-reference) in the
contiguous U.S. with flow data from 1941 to 2006. Reference streams have not
been substantially affected by dams and diversions; non-reference streams may or
may not have been affected in this way.
bBased on the annual range of 1-day flows. For each stream site, the median
variability was determined over a rolling 5-year window, then compared against the
baseline. Results are plotted at the midpoint of each window. For example, the value
for 2002-2006 is plotted at the year 2004.
Data source: Heinz Center, 2007
MRLC Consortium. 2007. National Land Cover Database
2001 (NLCD 2001). Accessed 2007.
Poff, N.L., and J.D. Allan. 1995. Functional organization
of stream fish assemblages in relation to hydrologic vari-
ability. Ecology 76:606-627.
Robinson, C.T., K. Tockner, andJ.V. Ward. 2002. The fauna
of dynamic riverine landscapes. Freshwater Biol. 47:661-677.
USGS (United States Geological Survey). 2007. National
Water Information System. Accessed 2007.
47
-------
INDICATO
High and Low Stream Flows
Exhibit 3-4. Percent of grassland/shrubland
streams in the contiguous U.S. experiencing
periods of no flow, by ecoregion, 1961-2006ab
100
I 60
Ol
~ 40
I
^ 20
1960
1970
1980 1990
Year
2000
2010
Ecoregion:
— California/Mediterranean
Desert/shrub
— Grassland/steppe
— All three of these ecoregions
Ecoregion divisions
.Desert/shrub
California/ Grassland/
Mediterranean steppe
Coverage: 280 stream gauging sites in watersheds containing 50
percent or greater grass/shrub cover, with flow data from 1941 to
2006. Grass/shrub cover refers to classes 52 and 71 of the 2001
National Land Cover Database (NLCD).
bStreams were classified based on annual data, then the
percentage of streams in each category was averaged over a
rolling 5-year window. Results are plotted at the midpoint of each
window. For example, the average for 2002-2006 is plotted at the
year 2004.
:Ecoregions based on Bailey (1995).
Data source: Heinz Center, 2007
Exhibit 3-5. Changes in the maximum duration
of no-flow periods in intermittent grassland/
shrubland streams of the contiguous U.S.,
1961-2006, compared with 1941-1960 baselineab
100
60
40
20
1960
1970
1980 1990
Year
2000
2010
Coverage: 163 stream gauging sites in
watersheds containing 50 percent or greater
grass/shrub cover, with flow data from
1941 to 2006 and at least one no-flow day
during this period. Grass/shrub cover refers
to classes 52 and 71 of the 2001 National
Land Cover Database (NLCD).
bFor each stream site, the duration of the maximum no-flow period
in each year was averaged over a rolling 5-year window. Results
are plotted at the midpoint of each window. For example, the value
for 2002-2006 is plotted at the year 2004.
:A substantial increase means the no-flow period was more than 14
days longerthan the average duration during the 1941-1960
baseline period; a substantial decrease means the no-flow period
was more than 14 days shorter.
Data source: Heinz Center, 2007
48
-------
INDICATO
Nitrogen and Phosphorus Loads in Large Rivers
Nitrogen is a critical nutrient for plants and animals,
and terrestrial ecosystems and head-water streams have
a considerable ability to capture nitrogen or to reduce it
to N2 gas though the process of denitrification. Nitrogen
cycling and retention is thus one of the most important
functions of ecosystems (Vitousek et al., 2002). When
loads of nitrogen from fertilizer, septic tanks, and atmo-
spheric deposition exceed the capacity of terrestrial systems
(including croplands), the excess may enter surface waters,
•where it may have "cascading" harmful effects as it moves
downstream to coastal ecosystems (Galloway and Cowl-
ing, 2002). Other sources of excess nitrogen include direct
discharges from storm water or treated wastewater. This
indicator specifically focuses on nitrate, which is one of the
most bioavailable forms of nitrogen in bodies of water.
Phosphorus is a critical nutrient for all forms of life,
but like nitrogen, phosphorus that enters the environment
from anthropogenic sources may exceed the needs and
capacity of the terrestrial ecosystem. As a result, excess
phosphorus may enter lakes and streams. Because phospho-
rus is often the limiting nutrient in these bodies of water,
an excess may contribute to unsightly algal blooms, which
cause taste and odor problems and deplete oxygen needed
by fish and other aquatic species. In some cases, excess
phosphorus can combine with excess nitrogen to exacer-
bate algal blooms (i.e., in situations where algal growth is
co-limited by both nutrients), although excess nitrogen
usually has a larger effect downstream in coastal -waters.
The most common sources of phosphorus in rivers are fer-
tilizer and wastewater, including storm water and treated
wastewater discharged directly into the river. In most
•watersheds, the atmosphere is not an important source or
sink for phosphorus.
This indicator tracks trends in nitrate and phosphorus
loads carried by four of the largest rivers in the United
States: the Mississippi, Columbia, St. Lawrence, and
Susquehanna. While not inclusive of the entire nation,
these four rivers account for approximately 55 percent of
all fresh\vater flow entering the ocean from the contiguous
48 states, and have a broad geographical distribution. This
indicator relies on stream flow and water- quality data col-
lected by the U.S. Geological Survey (USGS), which has
monitored nutrient export from the Mississippi River since
the mid-1950s and from the Susquehanna, St. Lawrence,
and Columbia Rivers since the 1970s. Data were collected
near the mouth of each river except the St. Lawrence,
•which \vas sampled near the point where it leaves the
United States.
At the sites for which data are included in this indica-
tor, USGS recorded daily water levels and volumetric
discharge using permanent stream gauges. Water quality
samples were collected at least quarterly over the period
of interest, in some cases up to 15 times per year. USGS
Exhibit 3-10. Nitrate loads in four major
U.S. rivers, 1955-20043
2,500
2,000
1,500
& 1,000
0
1950 1960 1970 1980 1990 2000 2010
Year
150
100
50
Detail
ol—
1970
aMost measurements include
nitrate plus nitrite, but because
concentrations of nitrite are
typically insignificant relative to
nitrate, this mixture is simply
called "nitrate."
Data source: USGS, 2007a
1980 1990 2000 2010
Year
Areas drained by these
four rivers
Columbia
St. Lawrence
Mississippi
L •/
Susquehanna
calculated annual nitrogen load from these data using
regression models relating nitrogen concentration to dis-
charge, day-of-year (to capture seasonal effects), and time
(to capture any trend over the period). These models were
used to make daily estimates of concentrations, which were
multiplied by the daily flow to calculate the daily nutrient
load (Aulenbach, 2006; Heinz Center, 2005). Because data
on forms of nitrogen other than nitrate and nitrite are not
as prevalent in the historical record, this indicator only uses
49
-------
INDICATO
Nitrogen and Phosphorus Loads in Large Rivers
Exhibit 3-11. Total phosphorus loads in four
major U.S. rivers, 1971-2004
250
200
150
a. 100
r so
2010
2010
Areas drained by these
four rivers
Columbia
St. Lawrence
Mississippi
Susquehanna
Data source: USGS, 2007a
measurements of nitrate plus nitrite. As nitrite concentra-
tions are typically very small relative to nitrate, this mix-
ture is simply referred to as nitrate.
What the Data Show
The Mississippi River, which drains more than 40 percent
of the area of the contiguous 48 states, carries roughly 15
times more nitrate than any other U.S. river. Nitrate load
in the Mississippi increased noticeably over much of the last
half-century, rising from 200,000-500,000 tons per year in
the 1950s and 1960s to an average of about 1,000,000 tons
per year during the 1980s and 1990s (Exhibit 3-10). Large
year-to-year fluctuations are also evident. The Mississippi
drains the agricultural center of the nation and contains
a large percentage of the growing population, so it may
not be surprising that the -watershed has not been able to
assimilate all the nitrogen from sources such as crop and
lawn applications, animal manure and human wastes, and
atmospheric deposition (e.g., Rabalais and Turner, 2001).
The Columbia River's nitrate load increased to almost
twice its historical loads during the later half of the 1990s,
but by the last year of record (2002), the nitrate load had
returned to levels similar to those seen in the late 1970s
(Exhibit 3-10). The St. Lawrence River showed an overall
upward trend in nitrate load over the period of record,
•while the Susquehanna does not appear to have shown an
appreciable trend in either direction. Over the period of
record, the Columbia and St. Lawrence carried an average
of 67,000 and 66,000 tons of nitrate per year, respectively,
•while the Susquehanna averaged 46,000 tons. By compari-
son, the Mississippi carried an average of 772,000 tons per
year over its period of record.
The total phosphorus load decreased in the St. Lawrence
and Susquehanna Rivers over the period of record (Exhibit
3-11). There is no obvious trend in the Mississippi and
Columbia Rivers, and the year-to-year variability is quite
large. Nitrogen and phosphorus loads tend to be substan-
tially higher during years of high precipitation, because of
increased erosion and transport of the nutrients to stream
channels (Smith et al, 2003). Over the full period of
record, average annual phosphorus loads for the Mississippi,
Columbia, St. Lawrence, and Susquehanna were 138,000;
11,000; 6,000; and 3,000 tons, respectively.
Indicator Limitations
• The indicator does not include data from numerous
coastal -watersheds whose human populations are rapidly
increasing (e.g., Valigura et al., 2000).
• It does not include smaller -watersheds in geologically
sensitive areas, -whose ability to retain nitrogen might be
affected by acid deposition (e.g., Evans et al., 2000).
50
-------
INDICATO
Nitrogen and Phosphorus Loads in Large Rivers
• It does not include forms of nitrogen other than nitrate.
Although nitrate is one of the most bioavailable forms of
nitrogen, other forms may constitute a substantial portion
of the nitrogen load. Historically, nitrate data are more
extensive than data on other forms of nitrogen.
• Not all forms of phosphorus included in the total phos-
phorus loads are equally capable of causing algal blooms.
Data Sources
Data were compiled for EPA by USGS (USGS, 2007a),
•which provided a similar analysis to the Heinz Center for
its updated report. Nutrient loads for the Columbia, St.
Lawrence, and Susquehanna were originally reported in
Aulenbach (2006); portions of the Mississippi analysis were
previously published in Goolsby et al. (1999), while other
portions have not yet been published. Underlying nutrient
sampling and daily stream flow data can be obtained from
USGS's public databases (USGS, 2007b,c).
References
Aulenbach, B.T. 2006. Annual dissolved nitrite plus nitrate
and total phosphorus loads for Susquehanna, St. Lawrence,
Mississippi-Atchafalaya, and Columbia River Basins,
1968-2004. USGS Open File Report 06-1087.
Evans, C.D., A.Jenkins, and R.F. Wright. 2000. Surface
•water acidification in the South Pennines I. Current status
and spatial variability. Environ. Pollut. 109(1):11-20.
Galloway, J., andE. Cowling. 2002. Reactive nitrogen and
the \vorld: 200 years of change. Ambio 31:64-71.
Goolsby, D.A., W.A. Battaglm, G.B. Lawrence, R.S.
Artz, B.T. Aulenbach, R.P. Hooper, D.R. Keeney, and
GJ. Stensland. 1999. Flux and sources of nutrients in
the Mississippi-Atchafalaya River Basin—topic 3 report
for the integrated assessment on hypoxia in the Gulf of
Mexico. NOAA Coastal Ocean Program Decision Analy-
sis Series No. 17.
Heinz Center (The H. John Heinz III Center for Science,
Economics, and the Environment). 2005. The state of the
nation's ecosystems: Measuring the lands, -waters, and living
resources of the United States. New York, NY: Cambridge
University Press. Web update 2005:
Rabalais, N.N., and R.E. Turner, eds. 2001. Coastal
hypoxia: Consequences for living resources and ecosys-
tems. Coastal and estuarine studies 58. Washington, DC:
American Geophysical Union.
Smith, S.V., D.P. Swaney, L. Talaue-McManus, J.D.
Bartley, P.T Sandhei, CJ. McLaughlm, V.C. Dupra,
C.J. Grassland, RW. Buddemeier, B.A. Maxwell, and F.
Wulff. 2003. Humans, hydrology, and the distribution
of inorganic nutrient loading to the ocean. BioScience
53:235-245.
USGS (United States Geological Survey). 2007a. Data pro-
vided to ERG (an EPA contractor) by Nancy Baker, USGS.
September 12, 2007.
USGS. 2007b. National Stream Quality Accounting Net-
work (NASQAN) data. Accessed 2007.
USGS. 2007c. National Water Information System.
Accessed 2007.
Valigura, R., R. Alexander, M. Castro, T Meyers, H.
Paerl, P. Stacey, and R. Turner, eds. 2000. Nitrogen load-
ing in coastal -water bodies—an atmospheric perspective.
Washington, DC: American Geophysical Union.
Vitousek, P., H. Mooney, L. Olander, and S. Allison. 2002.
Nitrogen and nature. Ambio 31:97-101.
51
-------
Benthic Macroinvertebrates in Wadeable Streams
Freshwater benthic macroinvertebrate communities are
composed primarily of insect larvae, mollusks, and
•worms. They are an essential link in the aquatic food web,
providing food for fish and consuming algae and aquatic
vegetation (U.S. EPA, 2006b). The presence and distri-
bution of macroinvertebrates in streams can vary across
geographic locations based on elevation, stream gradient,
and substrate (Barbour et al., 1999). These organisms are
sensitive to disturbances in stream chemistry and physical
habitat, both in the stream channel and along the riparian
zone, and alterations to the physical habitat or water chem-
istry of the stream can have direct and indirect impacts on
their community structure. Because of their relatively long
life cycles (approximately 1 year) and limited migration,
benthic macroinvertebrates are particularly susceptible to
site-specific stressors (Barbour et al., 1999).
This indicator is based on data collected for EPA's
Wadeable Streams Assessment (WSA). Wadeable streams
are streams, creeks, and small rivers that are shallow
enough to be sampled using methods that involve wading
into the water. They typically include waters classified as
1st through 4th order in the Strahler Stream Order classi-
fication system (Strahler, 1952). Between 2000 and 2004,
crews sampled 1,392 sites throughout the contiguous U.S.
using standardized methods (U.S. EPA, 2004a,b). Sites
•were sampled between mid-April and mid-November. At
each site, a composite bottom sample was collected from
eleven equally spaced transects within the sample reach.
The WSA is based on a probabilistic design, so results from
the sample sites can be used to make statistically valid state-
ments about the percentage of wadeable stream miles that
fall above or below reference values for the indicator.
For this analysis, the 48 contiguous states were divided
into nine broad ecoregions (U.S. EPA, 2006b), which were
defined by the WSA based on groupings of EPA Level III
ecoregions (Omermk, 1987; U.S. EPA, 2007). Benthic
community condition was determined using two dif-
ferent approaches, each reflecting a distinct aspect of the
indicator: an Index of Biological Integrity (IBI) and an
observed/expected (O/E) predictive model.
The IBI is an index that reduces complex information
about community structure into a simple numerical value
based on measures of taxonomic richness (number of taxa);
taxonomic composition (e.g., insects vs. non-insects); taxo-
nomic diversity; feeding groups (e.g., shredders, scrapers,
or predators); habits (e.g., burrowing, clinging, or climbing
taxa); and tolerance to stressors. Separate metrics were used
for each of these categories in the nine WSA ecoregions,
based on their ability to best discriminate among streams.
Each metric was scaled against the 5th-95th percentiles for
the streams in each region to create an overall IBI, whose
value ranges from 0 to 100 (Stoddard et al., 2005).
Exhibit 3-13. Index of Biological Integrity (IBI) for
benthic macroinvertebrates in wadeable streams
of the contiguous U.S., by region, 2000-2004ab
IBI score:
Least
disturbed
Moderately
disturbed
Most
disturbed
Not assessed/
no data
Percent of stream miles in each category:
Eastern
Highlands
18.2
20.4
51.8
9.5
Plains and
Lowlands
West
All U.S.
29.0
29.0
40.0
45.1
25.9
27.4
28.2
24.9
41.9
2.0
1.7
5.0
aRegions based on groupings of
EPA Level III ecoregions
(Omernik, 1987; U.S. EPA,
2007).
bTotalsmaynotaddto100%
due to rounding.
Data source: U.S. EPA, 2006b
Regions
Eastern
Highlands
Plains and Lowlands
Once the overall IBI was established, a set of relatively
undisturbed sites was selected in order to determine the
range of IBI scores that would be expected among "least
disturbed" sites. A separate reference distribution was
developed for each ecoregion. Next, the IBI score for
every sampled site was compared to the distribution of IBI
scores among the ecoregion's reference sites. If a site's IBI
score \vas below the 5th percentile of the regional reference
distribution, the site was classified as "most disturbed."
This threshold was used because it offers a high degree of
confidence that the observed condition is statistically differ-
ent from the "least disturbed" reference condition. Streams
•with IBI scores above the 25th percentile of the reference
range were labeled "least disturbed," indicating a high
probability that they are similar to the relatively undis-
turbed reference sites. Streams falling between the 5th and
25th percentiles were classified as "moderately disturbed." In
addition to national totals, this indicator displays IBI scores
for three broad regions, which are composed of multiple
WSA ecoregions and which share major climate and land-
form characteristics (U.S. EPA, 2006b).
The O/E predictive model compares the actual number
of macroinvertebrate taxa observed at each WSA site (O)
•with the number expected (E) to be found at a site that is
52
-------
Benthic Macroinvertebrates in Wadeable Streams
in minimally disturbed condition (Armitage, 1987). First,
reference sites were divided into several groups based on the
observed benthic assemblages, and the probability of observ-
ing each taxon in each group of sites was determined. Next,
a multivariate model was used to characterize each group of
reference sites in terms of their shared physical characteristics
(variables that are largely unaffected by human influence,
such as soil type, elevation, and latitude). This predictive
model then was applied to each test site to determine which
group(s) of reference sites it should be compared to. For each
test site, the "expected" probability of observing each taxon
\vas calculated as a -weighted average based on the probabil-
ity of observing that taxon in a particular group of reference
sites and the probability that the test site is part of that par-
ticular group of sites, based on physical characteristics. The
total "E" for the test site was generated by adding the prob-
abilities of observing each of the individual taxa. The actual
number of taxa collected at the site (O) was divided by "E"
to arrive at an O/E ratio (Hawkins et al., 2000; Hawkins
and Carlisle, 2001). An O/E of 1.0 means the site's taxa rich-
ness is equal to the average for the reference sites. Each tenth
of a point below 1 suggests a 10 percent loss of taxa.
What the Data Show
Based on the IBI, slightly more than one-quarter of-wade-
able stream miles nation-wide (28.2 percent) -were classified as
"least disturbed" -with respect to benthic macroinvertebrate
condition, -while 41.9 percent -were in the "most disturbed"
category (Exhibit 3-13). Of the three major stream regions
in the nation (see the inset map, Exhibit 3-13), the eastern
highlands had the lowest percentage of "least disturbed"
stream miles (18.2 percent), -while the -western region had
the highest percentage (45.1 percent).
Because there are no agreed-upon thresholds for the O/E
model, the results are presented in 20 percent increments of
taxa losses for the contiguous 48 states (Exhibit 3-14). Nearly
40 percent (38.6 percent) of wadeable stream miles have lost
more than 20 percent of their macroinvertebrate taxa, com-
pared to comparable minimally disturbed reference sites, and
8.3 percent of stream miles have lost more than 60 percent
of their macroinvertebrate taxa.
Indicator Limitations
• Although the probability sampling design results in
unbiased estimates for the IBI and O/E in wadeable
streams during the April-November index period, values
may be different during other seasons.
• Reference conditions for the IBI and O/E vary from one
ecoregion to another in both number and quality, -which
limits the degree of ecoregional resolution at -which this
indicator can be calculated.
Exhibit 3-14. Percent loss of benthic
macroinvertebrate taxa in wadeable streams of
the contiguous U.S., relative to the number of
expected taxa, 2000-2004ab
100
60
40
20
1.8
7.4
28.8
25.3
18.4
11.9
^ 6.5
D Not assessed/no data
D No loss
D 0-20% loss
• 20-40% loss
D 40-60% loss
D 60-80% loss
D More than 80% loss
aNumber of expected taxa based
on minimally disturbed
reference sites.
bTotals may not add to 100% due
to rounding.
Data source: U.S. EPA, 2005
• Because "E" is subject to both model error and sam-
pling error, O/E values near 1.0 (above or below) do not
necessarily imply a gain or loss of species relative to the
reference conditions.
• Trend data are unavailable because this is the first time
that a survey on this broad scale has been conducted, and
the survey design does not allow trends to be calculated
•within a single sampling period (2000-2004). These data
•will serve as a baseline for future surveys.
Data Sources
The results shown in Exhibit 3-13 -were previously published
in EPAs 2006 Wadeable Streams Assessment (WSA) report
(U.S. EPA, 2006b). The data in Exhibit 3-14 are based on
frequency distributions provided by the WSA program (U.S.
EPA, 2005) (U.S. EPA [2006b] also presents results from the
O/E analysis, but using different categories). Data from indi-
vidual stream sites can be obtained from EPA's STORET
database (U.S. EPA, 2006a) (http://-www.epa.gov/o-wo-w/
streamsurvey/web_data.html).
References
Armitage, D. 1987. The prediction of the macroinver-
tebrate fauna of unpolluted running-water sites in Great
Britain using environmental data. Fresh-water Biol.
17:41-52.
Barbour, M.T., J. Gerritson, B.D. Snyder, andJ.B. Strib-
ling. 1999. Rapid bioassessment protocols for use in streams
53
-------
Benthic Macroinvertebrates in Wadeable Streams
and wadeable rivers: Periphyton, benthic macroinverte-
brates and fish. Second edition. EPA/841/B-99/002. Wash-
ington, DC: U.S. Environmental Protection Agency.
Hawkins, C.P., and D.M. Carlisle. 2001. Use of predictive
models for assessing the biological integrity of wetlands and
other aquatic habitats. In: Rader, R.B., and D.P. Batzer,
eds. Bioassessment and management of North American
wetlands. New York, NY: John Wiley & Sons. pp. 59-83.
Hawkins, C.P., R.H. Norris, J.N. Hogue, andJW. Fem-
inella. 2000. Development and evaluation of predictive
models for measuring the biological integrity of streams.
Ecol. Appl. 10:1456-1477.
Omernik, J.M. 1987. Ecoregions of the conterminous
United States. Map (scale 1:7,500,000). Ann. Assoc. Am.
Geog. 77(1):118-125.
Stoddard, J., D.V. Peck, S.G. Paulsen, J. Van Sickle, C.P.
Hawkins, A.T. Herlihy, R.M. Hughes, F. Wright, PR.
Kaufmann, D.P. Larsen, G. Lomnicky, A.R. Olsen,
S.A. Peterson, P.L. Rmgold, and T.R. Whittier. 2005.
An ecological assessment of western streams and rivers.
EPA/620/R-05/005. Washington, DC: U.S. Environmen-
tal Protection Agency,
Strahler, A.N. 1952. Dynamic basis of geomorphology.
Geol. Soc. Am. Bull. 63:923-938.
U.S. EPA (United States Environmental Protection Agency).
2007. Level III ecoregions of the conterminous United
States. Accessed November 2007.
U.S. EPA. 2006a. Data from the Wadeable Streams
Assessment. Accessed 2006.
U.S. EPA. 2006b. Wadeable Streams Assessment: A collab-
orative survey of the nation's streams. EPA/841/B-06/002.
U.S. EPA. 2005. Data provided to ERG (an EPA contrac-
tor) by Susan Holdsworth, EPA. December 2005.
U.S. EPA. 2004a. Wadeable streams assessment: Benthic
laboratory methods. EPA/841/B-04/007.
U.S. EPA. 2004b. Wadeable streams assessment: Field opera-
tions manual. EPA/841/B-04/004.
54
-------
INDICATO
Trophic State of Coastal Waters
While the presence of many water pollutants can lead
to decreases in coastal water quality, four interlinked
components related to trophic state are especially criti-
cal: nutrients (nitrogen and phosphorus), chlorophyll-a,
dissolved oxygen, and water clarity. "Trophic state"
generally refers to aspects of aquatic systems associated
•with the growth of algae, decreasing water transparency,
and low oxygen levels in the lower water column that
can harm fish and other aquatic life. Nitrogen is usually
the most important limiting nutrient in estuaries, driv-
ing large increases of microscopic phytoplankton called
"algal blooms" or increases of large aquatic bottom plants,
but phosphorus can become limiting in coastal systems
if nitrogen is abundant in a bioavailable form (U.S. EPA,
2003). Nitrogen and phosphorus can come from point
sources, such as wastewater treatment plants and indus-
trial effluents, and nonpoint sources, such as runoff from
farms, over-fertilized lawns, leaking septic systems, and
atmospheric deposition. Chlorophyll-a is a surrogate
measure of phytoplankton abundance in the water col-
umn. Chlorophyll-a levels are increased by nutrients and
decreased by filtering organisms (e.g., clams, mussels, or
oysters). High concentrations of chlorophyll-a indicate
overproduction of algae, which can lead to surface scums,
fish kills, and noxious odors (U.S. EPA, 2004). Low dis-
solved oxygen levels and decreased clarity caused by algal
blooms or the decay of organic matter from the -water-
shed are stressful to estuarine organisms. Reduced water
clarity (usually measured as the amount and type of light
penetrating water to a depth of 1 meter) can be caused
by algal blooms, sediment inputs from the -watershed, or
storm-related events that cause resuspension of sediments,
and can impair the normal growth of algae and other sub-
merged aquatic vegetation.
This indicator, developed as part of EPA's Coastal
Condition Report, is based on an index constructed from
probabilistic survey data on five components: dissolved
inorganic nitrogen, dissolved inorganic phosphorus,
chlorophyll-a, daytime dissolved oxygen in bottom or
near-bottom -waters (-where benthic life is most likely to be
affected), and water clarity (U.S. EPA, 2004). The survey,
part of EPA's National Coastal Assessment (NCA), was
designed to provide a national picture of water quality by
sampling sites in estuarine -waters throughout the contigu-
ous 48 states and Puerto Rico. Each site -was sampled once
during the 1997-2000 period, -within an index period from
July to September. The indicator reflects average condition
during this index period.
Key factors like sediment load, mixing processes, and
ecosystem sensitivity naturally vary across biogeographic
regions and even among estuaries -within regions. Thus,
reference guidelines for nutrients, -water clarity, and chloro-
phyll-a -were established based on variable expectations for
Exhibit
contiguc
Region,
Region 1
Region 2
Regions
Region 4
Region 6
Region 9
Region 10
All U.S.b
3-21. Coastal water quality index for the
ius U.S. and Puerto Rico, by EPA
1997-20003
Water quality:
High Moderate Low Unsampled
'ercent of estuarine area in each category:
71 20 ||l 8
9 48 8 35
8 52 36 4
46 46 8
38 55 7
23 62 15
29 70 ||l
40 49 I 11 I
Coverage: Estuarine waters of the EPA Regions
contiguous 48 states and Puerto Rico. ^ ~ O
Does not include the hypoxic zone in
offshore Gulf Coast waters.
bU.S. figures reflect the total sampled
area. Unsampled areas were not tt(E>
included in the calculation. \A^~\ I©H
Data source: U.S. EPA, 2004, 2005a
conditions in different biogeographic regions. For example,
due to Pacific up-welling during the summer, higher nutri-
ent and chlorophyll-a concentrations are expected in West
Coast estuaries than in other estuaries. Water clarity refer-
ence guidelines are lower for estuaries that support sea-
grass than for naturally turbid estuaries. A single national
reference range of 2-5 milligrams per liter (mg/L) was
used for dissolved oxygen, because concentrations below 2
mg/L are almost always harmful to many forms of aquatic
life and concentrations above 5 mg/L seldom are (Diaz and
Rosenberg, 1995; U.S. EPA, 2000). The process of clas-
sifying individual sites varies by region and is described in
detail, along -with the regional reference conditions, in U.S.
EPA (2004).
The overall -water quality index is a compilation of the
five components. For each site, the index is rated high if
none of the five components received a score that -would
55
-------
INDICATO
Trophic State of Coastal Waters
Exhibit
waters c
by EPA
Region 1
Region 2
Regions
Region 4
Region 6
Region 9
Region 10
All U.S.d
Coverage:
contiguous
bThis indica
inorganic r
the sum ol
ammonia.
cTotals may
rounding.
dU.S. figure
Iwere not ir
Data soun
3-22. Nitrogen concentrations in coastal
if the contiguous U.S. and Puerto Rico,
Region, 1997-2000abc
Nitrogen concentration:
Low Moderate High Unsampled
'ercent of estuarine area in each category:
74 17 9
27 23 15 34
63 14 16 7
88 12
84 13
88 12
100
82 13
Estuarine waters of the EPA Regions
48 states and Puerto Rico. (ftfTiri
tor measures dissolved
itrogen (DIN), which is
nitrate, nitrite, and
*"\ >'~^^
s®
notadd to 100% due to Q>J ||
3 reflect the total sampled area. Unsampled areas
eluded in the calculation.
e: U.S. EPA, 2004, 2005a
3
5
f
be considered environmentally unfavorable (high nitro-
gen, phosphorus, or chlorophyll-a levels or low dissolved
oxygen or water clarity), and no more than one compo-
nent was rated moderate. Overall water quality is low if
more than two components received the most unfavorable
rating. All other sites receive a moderate index score. If
two or more components are missing, and the avail-
able components do not suggest a moderate or low index
rating, the site is classified as "unsampled." Data from
the individual sites were expanded from the probability
sample to provide unbiased estimates of the water quality
index and each of its components for each EPA Region.
Results were also aggregated and -weighted by estuarine
area for the entire nation.
Exhibit 3-23. Phosphorus concentrations in
coastal waters of the contiguous U.S. and
Puerto Rico, by EPA Region, 1997-2000abc
Phosphorus concentration:
Low
Moderate
High
Unsampled
Percent of estuarine area in each category:
Region 1
Region 2
Regions
Region 4
Region 6
Region 9
Region 10
All U.S.d
58
16
32 ||l
41
9
34
58
29 66
66
24 10
10
49
37 15
2
46
52
5 95
53
38
9
EPA Regions
Coverage: Estuarine waters of the
contiguous 48 states and Puerto Rico.
bThis indicator measures dissolved
inorganic phosphorus (DIP), which
equals orthophosphate.
Totals may not add to 100% due to
rounding.
dU.S. figures reflect the total
sampled area. Unsampled areas were not included in the
calculation.
Data source: U.S. EPA, 2004, 2005a
What the Data Show
According to the index, 40 percent of estuarine surface
area nation-wide exhibited high water quality over the
1997-2000 period, 11 percent had low water quality, and
the remaining 49 percent was rated moderate (Exhibit
3-21). Scores vary considerably among EPA Regions,
ranging from high water quality in 71 percent of estuarine
area in Region 1 to less than 10 percent in Regions 2 and
3. Only one EPA Region had low water quality in more
than 15 percent of its estuarine area (EPA Region 3, with
36 percent). These percentages do not include the Great
Lakes or the hypoxic zone in offshore Gulf Coast waters
(see the Hypoxia in Gulf of Mexico and Long Island
Sound indicator).
56
-------
INDICATO
Trophic State of Coastal Waters
Exhibit 3-24. Chlorophyll-a concentrations in
coastal waters of the contiguous U.S. and Puerto
Rico, by EPA Region, 1997-2000ab
Chlorophyll-a concentration:
Low
Moderate
High
Unsampled
88
6
6
24
13
33
31 7
54
37
27
57
6
10
53
41
6
Percent of estuarine area in each category:
Region 1
Region 2
Regions
Region 4
Region 6
Region 9
Region 10
All U.S.C
Coverage: Estuarine waters of the
contiguous 48 states and Puerto Rico.
"Totals may not add to 100% due to
rounding.
=U.S. figures reflect the total sampled
area. Unsampled areas were not
included in the calculation.
Data source: U.S. EPA, 2004, 2005a
87
13
80
20
51
41
8
EPA Regions
Nitrogen concentrations were low in 82 percent of
estuarine area and high in 5 percent nation-wide, and were
low in a majority of the estuarine area in all but one EPA
Region (Exhibit 3-22). Regions 2 and 3 had the largest
percentage of area with high concentrations (15 percent
and 16 percent, respectively); several other EPA Regions
had no areas with high concentrations.
Phosphorus concentrations were low in 53 percent of
estuarine area and high in 9 percent nation-wide (Exhibit
3-23). Region 9 had the largest proportion of area exceed-
ing reference conditions (52 percent), -while Region 10 had
the least (none).
Chlorophyll-a concentrations -were low in 51 per-
cent and high in 8 percent of estuarine area nation-wide
(Exhibit 3-24). Region 3 had the largest percentage of area
exceeding reference conditions (27 percent); all other EPA
Regions had 10 percent or less in this category.
Exhibit 3-25. Dissolved oxygen levels in
coastal waters of the contiguous U.S. and
Puerto Rico, by EPA Region, 1 997-2000 ab
Dissolved oxygen concentration:
High
(>5mg/L)
Moderate Low
(2-5mg/L) (<2mg/L) UnsamP|ed
Percent of estuarine area in each category:
Region 1
87 11
2
Region 2
66 21 13
Regions
53 21 21 4
Region 4
75 23
Region 6
84 16
Region 9
98
Region 10
67 31
2
2
2
All U.S.C
76 20 4
Coverage: Bottom- or near EPA Regions
bottom-water dissolved oxygen in flh "TjJ^ , ejft
estuarine waters of the contiguous
48 states and Puerto Rico. Does not
include the hypoxic zone in offshore
Gulf Coast waters.
bTotals may not add to 1
rounding.
CU.S. figures reflect the 1
not included in the calc
Data source: U.S. EPA,
00% due to ft® r
. E L_
otal sampled area. Unsampled areas were
ulation.
2004, 2005a
Bottom-water dissolved oxygen -was above 5 mg/L
in over three-fourths of the nation's estuarine area and
below 2 mg/L in only 4 percent (Exhibit 3-25). While
effects vary -with temperature and salinity, as a general
rule, concentrations of dissolved oxygen above 5 mg/L are
considered supportive of marine life, concentrations below
5 mg/L are potentially harmful, and concentrations below
2 mg/L—a common threshold for hypoxia—are associated
•with a -wider range of harmful effects (e.g., some juvenile
fish and crustaceans that cannot leave the area may die).
Region 3 had the greatest proportion of estuarine area
•with low dissolved oxygen (21 percent), -while four EPA
Regions had no area below 2 mg/L.
57
-------
INDICATO
Trophic State of Coastal Waters
Exhibit 3-26. Water clarity in coastal waters of
the contiguous U.S. and Puerto Rico, by EPA
Region, 1997-2000ab
Water clarity:
High
Moderate
Low
Unsampled
Percent of estuarine area in each category:
Region 1
Region 2
Region3
Region 4
Region 6
Region 9
Region 10
All U.S.C
95
82
10
38
14
43 5
66
47
12
22
65
44
19
31
11
24
18
62
38
13
25
Coverage: Estuarine waters of the
contiguous 48 states and Puerto Rico.
"Totals may not add to 100% due to
rounding.
=U.S. figures reflect the total sampled
area. Unsampled areas were not
included in the calculation.
Data source: U.S. EPA, 2004, 2005a
EPA Regions
Water clarity exceeded reference conditions (i.e., higher
clarity) in 62 percent of the nation's estuarine area, while
low water clarity was observed in 25 percent of estuarine
area (Exhibit 3-26). Region 3 had the largest proportion of
area with low clarity (43 percent), while Region 1 had the
smallest (none).
Indicator Limitations
• The coastal areas of Hawaii and a portion of Alaska have
been sampled, but the data had not yet been assessed
at the time this indicator was compiled. Data are also
not available for the U.S. Virgin Islands and the Pacific
territories.
• Trend data are not yet available for this indicator.
Because of differences in methodology, the data pre-
sented here are not comparable with data that appeared
in EPA's first National Coastal Condition Report. The
data presented here will serve as a baseline for future
surveys.
• The NCA surveys measure dissolved oxygen conditions
only in estuarine waters and do not include observa-
tions of dissolved oxygen concentrations in offshore
coastal shelf \vaters, such as the hypoxic zone in Gulf
of Mexico shelf \vaters.
• At each sample location, the components of this indica-
tor may have a high level of temporal variability. This
survey is intended to characterize the typical distribution
of-water quality conditions in coastal -waters during an
index period from July through September. It does not
consistently identify the "-worst-case" condition for sites
experiencing occasional or infrequent hypoxia, nutrient
enrichment, or decreased -water clarity at other times of
the year.
Data Sources
This indicator is based on an analysis published in EPA's
second National Coastal Condition Report (U.S. EPA,
2004). Summary data by EPA Region have not been pub-
lished, but -were provided by EPA's NCA program (U.S.
EPA, 2005a). Underlying sampling data are housed in
EPA's NCA database (U.S. EPA, 2005b) (http://www.epa.
gov/emap/nca/html/data/index.html).
References
Diaz, R.J., and R. Rosenberg. 1995. Marine benthic
hypoxia: A review of its ecological effects and the behav-
ioral responses of benthic macrofauna. Oceanogr. Mar.
Biol. Ann. Rev. 33:245-303.
U.S. EPA (United States Environmental Protection
Agency). 2005a. Data provided to ERG (an EPA contractor)
by Kevin Summers, EPA. September 2005.
U.S. EPA. 2005b. EMAP national coastal database.
Accessed 2005.
U.S. EPA. 2004. National coastal condition report II.
EPA/620/R-03/002.
U.S. EPA. 2003. Mid-Atlantic integrated assessment,
MAIA—estuaries 1997-98, summary report.
EPA/620/R-02/003.
U.S. EPA. 2000. Ambient aquatic life -water quality criteria
for dissolved oxygen (salt-water): Cape Cod to Cape Hat-
teras. EPA/822/R-00/12.
58
-------
INDICATOR
Coastal Sediment Quality
Contaminated sediments can pose an immediate threat to
benthic organisms and an eventual threat to entire estua-
rine ecosystems. Sediments can be resuspended by anthro-
pogenic activities, storms, or other natural events; as a result,
organisms in the water column can be exposed to contami-
nants, which may accumulate through the food web and
eventually pose health risks to humans (U.S. EPA, 2004a).
There are several ways to measure sediment quality.
Sediments can be assessed in terms of their toxicity to
specific organisms in bioassays, or in terms of the levels of
contaminants that are present. Sediment quality also can be
inferred by assessing the condition of benthic communities,
•which largely reflect the quality of the sediments in which
they live (although other stressors may be reflected as well).
To generate a more complete picture of sediment quality,
scientists frequently use several of these measures together.
This indicator presents data on sediment toxicity and
contaminant levels. The data are from probabilistic surveys
conducted as part of EPA's National Coastal Assessment
(NCA) and presented in EPA's second National Coastal
Condition Report (U.S. EPA, 2004b). The survey was
designed to provide a national picture of sediment quality
by sampling sites in estuarine waters throughout the contig-
uous 48 states and Puerto Rico. Each site was sampled once
during the 1997-2000 period, within an index period from
July to September. The indicator reflects average condi-
tion in each EPA Region during this index period. Results
•were also aggregated and -weighted by estuarine area for the
entire nation.
Sediment toxicity is typically determined using bioas-
says that expose test organisms to sediments and evaluate
their effects on the organisms' survival. For this indicator,
toxicity \vas determined using a 10-day static test on the
benthic amphipod Ampelisca abdita, which is commonly
used as a screening tool to identify sediments that pose
sufficient concern to -warrant further study. Sediments -were
classified as "potentially toxic" if the bioassays resulted in
greater than 20 percent mortality (a reference condition),
or "not likely toxic" if the bioassays resulted in 20 percent
mortality or less (U.S. EPA, 2004c).
Contaminant concentrations do not directly reflect
toxicity because toxicity also depends on contaminants'
bioavailability, -which is controlled by pH, particle size and
type, organic content, and other factors (e.g., mercury vs.
methylmercury). Contaminant concentrations are a use-
ful screening tool for toxicity, however, -when compared
•with concentrations known to cause particular effects on
benthic life. For this indicator, sediment samples -were
homogenized and analyzed for nearly 100 contaminants,
including 25 polycyclic aromatic hydrocarbons (PAHs),
22 polychlorinated biphenyls (PCBs), 25 pesticides, and 15
metals, using standard wet chemistry and mass spectros-
copy. The observed concentrations -were then compared
•with "effects range median" (ERM) values established
Exhibit 3-27. Sediment toxicity in coastal waters
of the contiguous U.S. and Puerto Rico, by EPA
Region, 1997-2000ab
Sediment toxicity:
Not likely toxic0
Potentially toxicd
Un samp led
Percent of estuarine area in each category:
Region 1
Region 2
Regions
Region 4
Region 6
Region 9
Region 10
All U.S.6
89.0
5.6
73.0
13.0 13.9
92.3
4.7
87.2
0.2 12.6
38.7
1.0
60.3
99.8
79.9
20.1
94.0
5.4
3.0
0.2
6.0
EPA Regions
Coverage: Estuarine waters of the
contiguous 48 states and Puerto
Rico.
"Totals may not add to 100% due to
rounding.
cNot likely toxic: Mortality of test
species = 20% or lower
Potentially toxic: Mortality of test
species > 20%
eU.S. figures reflect the total sampled area. Unsampled areas were
not included in the calculation.
Data source: U.S. EPA, 2004b, 2005a
through an extensive review of toxicity tests involving
benthic organisms, mostly Ampelisca (Long et al., 1995).
ERM values were available for 28 contaminants. For each
contaminant, the ERM represents the concentration at
•which there is a 50 percent likelihood of adverse effects to
an organism, based on experimental data. For this indi-
cator, a site \vas rated "potentially toxic" if one or more
contaminants exceeded an ERM value. In practice, about
25 percent of samples that exceed one ERM also cause
more than 20 percent mortality in the Ampelisca bioassay
(Long, 2000).
Benthic community condition also can be a useful indi-
cation of sediment quality, particularly in terms of chronic
or community effects that would not be captured in an
acute exposure bioassay. The NCA evaluated estuarine
59
-------
INDICATOR
Coastal Sediment Quality
sites for several aspects of benthic community condition,
and these results are presented as a separate ROE indicator
(Coastal Benthic Communities).
What the Data Show
Nationwide, 6 percent of coastal sediments were rated
"potentially toxic" based on the Ampelisca toxicity screen-
ing assay, although there was considerable variability from
one EPA Region to the next (Exhibit 3-27). In Region 9,
nearly 100 percent of estuarine area exhibited low sedi-
ment toxicity, while in some other EPA Regions, as much
as 20 percent of estuarine sediments were "potentially
toxic." Data for Region 6 are inconclusive because more
than half of the Region's estuarine area was not sampled.
Nationally, contaminants were present at "potentially
toxic" levels in 7 percent of estuarine sediments for which
contamination data were available (Exhibit 3-28). There was
considerable variability in sediment contamination from one
EPA Region to the next, with Region 4 showing the largest
proportion of estuarine area with sediments not likely to be
toxic (99.9 percent) and Region 2 showing the largest pro-
portion with "potentially toxic" sediments (24.4 percent).
Although the two figures suggest that a similar percent-
age of the nation's estuarine sediments are "potentially
toxic," the original data source reports very little correla-
tion between sites that caused more than 20 percent mor-
tality in the Ampelisca bioassay and sites where one or more
contaminants exceeded the ERM (U.S. EPA, 2004b). It
is not unusual to find a lack of correlation—particularly
in cases where sediment contaminants are neither highly
concentrated nor completely absent—in part because some
toxic chemicals may not be bioavailable, some may not be
lethal, and not all potentially toxic chemicals are ana-
lyzed (see O'Connor et al., 1998, and O'Connor and Paul,
2000). These results underscore the utility of a combined
approach to screen for potentially toxic sediments.
Indicator Limitations
• The coastal areas of Hawaii and a portion of Alaska have
been sampled, but the data had not yet been assessed at the
time this indicator was compiled. Data are also not avail-
able for the U.S. Virgin Islands and the Pacific territories.
• Trend data are not yet available for this indicator. Because
of differences in methodology, the data presented here
are not comparable with data that appeared in EPA's first
National Coastal Condition Report. The data presented
here will serve as a baseline for future surveys.
• Sample collection is limited to an index period from July
to September. It is not likely that contaminant levels vary
from season to season, however.
• The Ampelisca bioassay is a single-organism screening
tool, and the ERMs are general screening guidelines
based largely on toxicity data from Ampelisca. Thus,
Exhibit 3-28. Sediment contamination in
coastal waters of the contiguous U.S. and
Puerto Rico, by EPA Region, 1997-20003
Sediment contamination:
Not likely toxicb
Potentially toxic0
Unsampled
Percent of estuarine area in each category:
Region 1
Region 2
Regions
Region 4
Region 6
Region 9
Region 10
All U.S.d
90.0
4.6
5.4
62.0
24.4
13.6
88.2
9.4
2.4
99.9
0.1
81.7
18.3
92.8
98.1
93.0
1.9
17.0
"Coverage: Estuarine waters of the ^ EPA Regions
contiguous 48 states and Puerto
Rico.
bNot likely toxic: No contaminants
above effects range median (ERM)
Potentially toxic: One or more
contaminants above effects range
median (ERM)
dU.S. figures reflect the total sampled area. Unsampled areas were
not included in the calculation.
Data source: U.S. EPA, 2004b, 2005a
these measures do not necessarily reflect the extent to
•which sediments may be toxic to the full range of biota
(including microbes and plants) that inhabit a particular
sampling location.
The Ampelisca bioassay tests only for short-term, not
long-term, exposure. Both screening tests characterize
sediments in terms of their effects on benthic organism
mortality. This indicator does not capture other effects
of sediment contaminants on benthic organisms, such as
disease, stress, and reproductive effects.
This indicator cannot be compared quantitatively with
indicators that use other types of contaminant guidelines.
For example, the Pesticides in Agricultural Streams indi-
cator uses thresholds intended to be protective of aquatic
life with a margin of safety, instead of thresholds shown
60
-------
INDICATOR
Coastal Sediment Quality
to cause biological effects (e.g., ERMs). The ERM
approach also is not directly comparable with other sedi-
ment contaminant approaches, such as EPA's equilibrium
partitioning (EqP) benchmarks.
Data Sources
This indicator is based on an analysis published in EPA's sec-
ond National Coastal Condition Report (U.S. EPA, 2004b).
Summary data by EPA Region have not been published,
but were provided by EPA's NCA program (U.S. EPA,
2005a). Underlying sampling data are housed in EPA's NCA
database (U.S. EPA, 2005b) (http://www.epa.gov/emap/
nca/htm I/data/index, html).
References
Long, E.R. 2000. Degraded sediment quality in U.S.
estuaries: A review of magnitude and ecological applica-
tions. Ecol. Appl. 10(2):338-349.
Long, E.R., D.D. MacDonald, L. Smith, and ED. Calder.
1995. Incidence of adverse biological effects within ranges of
chemical concentrations in marine and estuarine sediments.
Environ. Manage. 19:81-97.
O'Connor, T.P., K.D. Daskalakis, J.L. HylandJ.F. Paul,
andJ.K. Summers. 1998. Comparisons of sediment toxicity
•with predictions based on chemical guidelines. Environ.
Toxicol. Chem. 17(3):468-471.
O'Connor, T.P., andJ.F. Paul. 2000. Misfit between
sediment toxicity and chemistry. Mar. Pollut. Bull.
40(l):59-64.
U.S. EPA (United States Environmental Protection Agency).
2005a. Data provided to ERG (an EPA contractor) by Kevin
Summers, EPA. September 2005.
U.S. EPA. 2005b. EMAP national coastal database.
Accessed 2005.
U.S. EPA. 2004a. Contaminated sediment in water.
U.S. EPA. 2004b. National coastal condition report II.
EPA/620/R-03/002.
U.S. EPA. 2004c. The incidence and severity of sediment
contamination in surface waters of the United States—
national sediment quality survey. Second edition.
61
-------
INDICATOR
Coastal Benthic Communities
Benthic communities are largely composed of macro-
invertebrates, such as annelids, mollusks, and crusta-
ceans. These organisms inhabit the bottom substrates of
estuaries and play a vital role in maintaining sediment and
•water quality. They also are an important food source for
bottom-feeding fish, invertebrates, and birds. Communi-
ties of benthic organisms are important indicators of envi-
ronmental stress because they are particularly sensitive to
pollutant exposure (Holland et al., 1987). This sensitivity
arises from the close relationship between benthic organ-
isms and sediments—which can accumulate environmental
contaminants over time—and the fact that these organisms
are relatively immobile, which means they receive pro-
longed exposure to any contaminants in their immediate
habitat (Sanders et al., 1980; Nixon et al., 1986).
This indicator is based on a multi-metric benthic com-
munities index that reflects overall species diversity in
estuarine areas throughout the contiguous United States
(adjusted for salinity, if necessary) and, for some regions,
the presence of pollution-tolerant and pollution-sensitive
species (e.g., Weisberg et al., 1997; Engle and Summers,
1999; U.S. EPA, 2004). The benthic community condition
at each sample site is given a high score if the index exceeds
a particular threshold (e.g., has high diversity or populations
of many pollution-sensitive species), a low score if it falls
below the threshold conditions, and a moderate score if it
falls within the threshold range. The exact structure of the
index and the threshold values vary from one biogeographic
region to another, but comparisons between predicted and
observed scores based on expert judgment are used to ensure
that the classifications of sites from one region to another
are consistent (U.S. EPA, 2004). Data were collected using
probability samples, so the results from the sampling sites
provide unbiased estimates of the distribution of index scores
in estuaries throughout each region.
The data for this indicator are from probabilistic surveys
conducted as part of EPA's National Coastal Assessment
(NCA) and presented in EPA's second National Coastal
Condition Report (U.S. EPA, 2004). The survey was
designed to provide a national picture of coastal benthic
community condition by sampling sites in estuarine waters
throughout the contiguous 48 states and Puerto Rico. Each
site was sampled once during the 1997-2000 period, within
an index period from July to September. The indicator
reflects average condition in each EPA Region during this
index period. Results were also aggregated and -weighted
by estuarine area for the entire nation.
What the Data Show
Nationally, 70 percent of the sampled estuarine area had a
high benthic communities index score, with 13 percent in
the moderate range and 17 percent scoring low (Exhibit
3-29). Condition varied somewhat by EPA Region, with
high index scores ranging from 51 percent of the estuarine
Exhibit
index fo
by EPA
Region 1
Region 2
Regions
Region 4
Region 6
Region 9
Region 10
All U.S.b
3-29. Coastal benthic communities
rthe contiguous U.S. and Puerto Rico,
Region, 1 997-2000 a
Benthic community condition:
High Moderate Low Unsamplec
Percent of estuarine area in each category:
77 18 5
59 17 24
72 27
63 17 17 :
1
51 27 11 11
78 15 7
91 5
70 13 17
t
Coverage: Estuarine waters of the EPA Regions
contiguous 48 states and ejfb
bU.S. figures reflect the total
sampled area. Unsampled areas
were not included in the calculation. ta(£)
Data source: U.S. EPA, 2004, , <§>» ©—
2005a
area in Region 6 to 91 percent in Region 10. Region 3
had the largest proportion of estuarine area rated low (27
percent), while Region 10 had the lowest (4 percent). In
the figure, the portion of the estuarine area not represented
by the sample is noted for each Region.
The National Coastal Condition Report found that
many of the sites with low benthic community condition
also showed impaired water quality or sediment condi-
tion—\vhich is not surprising given the extent to -which
these stressors and effects are related. Of the 17 percent of
national estuarine area rated low on the benthic commu-
nities index, 38 percent also exhibited degraded sediment
quality, 9 percent exhibited degraded -water quality (U.S.
EPA, 2004), and 33 percent exhibited degraded quality of
both sediment and -water.
Indicator Limitations
• The coastal areas of Hawaii and a portion of Alaska have
been sampled, but the data had not yet been assessed at the
62
-------
INDICATOR
Coastal Benthic Communities
time this indicator was compiled. Data are also not avail-
able for the U.S. Virgin Islands and the Pacific territories.
• Trend data are not yet available for this indicator. Because
of differences in methodology, the data presented here
are not comparable with data that appeared in EPA's first
National Coastal Condition Report. The data presented
here will serve as a baseline for future surveys.
• Benthic indices for the Northeast, West, and Puerto
Rico do not yet include measures of pollution-tolerant or
pollution-sensitive species. Although species diversity has
the largest impact on index scores in the other regions,
index values could change in the future as these compo-
nents are added to the index values for these regions.
• Sample collection is limited to an index period from July
to September. Further, because benthic communities can
be strongly influenced by episodic events, trawling, or
climate perturbations, this indicator may not reflect the
full range of conditions that occur at each sampling loca-
tion throughout these months.
Data Sources
This indicator is based on an analysis published in EPA's
second National Coastal Condition Report (U.S. EPA,
2004). Summary data by EPA Region have not been pub-
lished, but were provided by EPA's NCA program (U.S.
EPA, 2005a). Underlying sampling data are housed in
EPA's NCA database (U.S. EPA, 2005b) (http://www.epa.
gov/emap/nca/html/data/index.html).
References
Engle, V.D., andJ.K. Summers. 1999. Refinement,
validation, and application of a benthic condition
index for northern Gulf of Mexico estuaries. Estuaries
22(3A):624-635.
Holland, A.F., A. Shaughnessy, andM.H. Heigel. 1987.
Long-term variation in mesohaline Chesapeake Bay ben-
thos: Spatial and temporal patterns. Estuaries 10:227-245.
Nixon, S.W., C.D. Hunt, andB.L. Nowicki. 1986. The
retention of nutrients (C, N, P), heavy metals (Mn, Cd,
Pb, Cu), and petroleum hydrocarbons by Narragansett
Bay. In: Lasserre, P., andJ.M. Martin, eds. Biogeochemi-
cal processes at the land-sea boundary. New York, NY:
Elsevier. pp. 99-122.
Sanders, H.L., J.F. Grassle, G.R. Hampson, L.S. Morse,
S. Gerner-Price, and C.C.Jones. 1980. Anatomy of an oil
spill: Long-term effects from the grounding of the barge
Florida off West Falmouth, Massachusetts. J. Mar. Res.
38:265-380.
U.S. EPA (United States Environmental Protection Agency).
2005a. Data provided to ERG (an EPA contractor) by Kevin
Summers, EPA. September 2005.
U.S. EPA. 2005b. EMAP national coastal database.
Accessed 2005.
U.S. EPA. 2004. National coastal condition report II.
EPA/620/R-03/002.
Weisberg, S.B., T.A. Ranasmghe, D.D. Dauer, L.C.
o" ' J o
Schnaffer, R.J. Diaz, andJ.B. Frithsen. 1997. An estuarine
benthic index of biotic integrity (B-IBI) for Chesapeake
Bay. Estuaries 20(1):149-158.
63
-------
INDICATOR
Coastal Fish Tissue Contaminants
Contaminants in fish not only affect the fish's own health
and ability to reproduce, but also affect the many spe-
cies that feed on them. Contaminants also may make fish
unsuitable for human consumption (U.S. EPA, 2000).
This indicator, derived from an indicator presented in
EPA's second National Coastal Condition Report (U.S.
EPA, 2004), is based on National Coastal Assessment (NCA)
fish tissue survey data from 653 estuarine sites throughout
the United States. The survey was designed to provide a
national picture of coastal fish tissue contaminants by sam-
pling sites in estuarine waters throughout the contiguous
48 states. Each site was sampled once during the 1997-2000
period, within an index period from July to September. The
indicator reflects average condition in each EPA Region
during this index period. Results were also aggregated and
•weighted by estuarine area for the entire nation.
Fish and shellfish analyzed in the survey included
Atlantic croaker, white perch, catfish, flounder, scup, blue
crab, lobster, shrimp, whiffs, mullet, tomcod, spot, -weak-
fish, halibut, sole, sculpins, sanddabs, bass, and sturgeon.
At each site, five to 10 whole-body fish samples were tested
for 90 contaminants. This indicator is based on data col-
lected from 1997 to 2000.
To assess risks to human health, contaminant concentra-
tions in fish tissue were compared with established EPA
guideline ranges for recreational fishers, which were avail-
able for 16 of the 90 analytes. These guideline ranges are
based on the consumption of four 8-ounce fish meals per
month, and generally reflect non-cancer risks (U.S. EPA,
2000, 2004). For most contaminants, this is done using
•whole-body concentrations; for mercury, which concen-
trates in the edible fillet portion of the fish, a factor of 3.0
•was used to correct whole-body concentrations in order to
approximate fillet concentrations. The 3.0 factor represents
the median value (range 1.5-5.0) found in the available
literature (Wmdom and Kendall, 1979; Mikac et al., 1985;
Schmidt and Brumbaugh, 1990; Kannan et al., 1998;
Canadian Council of Ministers of the Environment, 1999).
For this indicator, a site was given a high contamina-
tion score if one or more contaminants were present at a
concentration above the guideline ranges. A site was rated
moderate if one or more contaminants were within the
guideline ranges but none was in exceedance. Sites with
all contaminants below their guideline ranges were given a
low contamination score.
What the Data Show
Nationwide, 63 percent of sites showed low fish tissue
contamination, 15 percent had moderate contamination,
and 22 percent exhibited high contamination (Exhibit
3-38). Fish tissue contamination varied substantially from
one EPA Region to the next; for example, the percentage
of sites \vith low contamination ranged from 25 percent
(Region 1) to 83 percent (Region 4). Regions 2 and 9 had
Exhibit 3-38. Coastal fish tissue contaminants
in the contiguous U.S. by EPA Region,
1997-2000abc
Level of contamination:
Low
Moderate
High
25
38
37
Percent of estuarine sites in each category:
Region 1
Region 2
Region3
Region 4
Region 6
Region 9
Region 10
All U.S.
39
20
53
83
41
20 27
13 4
59
7 34
52 8 40
67 11
63 15
22
22
Coverage: Estuarine waters of the
contiguous 48 states.
bThis indicator is based on a
whole-body analysis of the fish. See
text for definitions of categories.
Totals may not add to 100% due to
rounding.
Data source: U.S. EPA, 2004,2005a
EPA Regions
the largest proportion of sites with high contamination (41
percent and 40 percent, respectively).
Data from EPA's National Coastal Database show that
nation-wide, PCBs were the contaminants most frequently
responsible for high fish tissue contamination, with 19
percent of sites above EPA guideline ranges (Exhibit 3-39).
Other chemicals present above EPA guideline ranges at
many sites were mercury in muscle tissue (18 percent of
sites), DDT (8 percent), and PAHs (3 percent) (Exhibit
3-39). Inorganic arsenic, selenium, chlordane, endosulfan,
endrin, heptachlor epoxide, hexachlorobenzene, lindane,
and mirex were below EPA guideline ranges for all fish
sampled in the NCA.
Indicator Limitations
• The indicator is limited to estuarine samples, and does not
include data from Louisiana, Florida, Puerto Rico, Alaska,
or Hawaii, which had not been assessed at the time this
64
-------
INDICATOR
Coastal Fish Tissue Contaminants
indicator was compiled. Some of these areas
(e.g., portions of Alaska) have now been sur-
veyed, and may be included in future indicators.
• The data are not broken out by trophic level of
the fish and shellfish species, which influences
bioaccumulation of contaminants.
• Whole-body contaminant concentrations
in fish overestimate the risk associated with
consuming only the fillet portion of the fish,
•with the exception of mercury and cadmium,
•which are generally underestimated.
• This indicator focuses on contaminants from
a human health risk perspective. No EPA
guidance criteria exist to assess the ecological
risk of \vhole-body contaminants in fish (U.S.
EPA, 2004).
• Some fish samples used in the survey were
non-market-size juveniles, which are known
to have lower contaminant levels than larger,
market-sized fish.
• Samples are collected during an index period
from July to September, and the indicator is
only representative of this time period. It is
unlikely, however, that contaminant levels
vary substantially from season to season.
• There are no trend data for this indicator.
In EPA's second National Coastal Condition
Report, fish tissue contaminants are charac-
terized by \vhole-body concentrations and
compared to EPA risk-based consumption
guideline ranges. For the first National Coastal
Condition Report, fish contaminants -were
measured as fillet concentrations and com-
pared to U.S. Food and Drug Administration
(FDA) criteria. The data presented here -will
serve as a baseline for future surveys, however.
Data Sources
This indicator is based on an analysis published
in EPA's second National Coastal Condition
Report (U.S. EPA, 2004). Summary data by
EPA Region and by contaminant have not been
published, but -were provided by EPA's NCA
program (U.S. EPA, 2005a). Underlying sam-
pling data are housed in EPA's NCA database
(U.S. EPA, 2005b) (http://www.epa.gov/emap/
nca/html/data/index.html).
Exhibit 3-39. Coastal fish tissue contaminant
concentrations in the contiguous U.S., compared with
health-based guidelines, 1997-2000abc
Percent of estuarine sites:
Contaminant
Below Within Exceeding
Guideline range guideline guideline guideline
(ppm) range range range
Arsenic (inorganic)1
Cadmium
Mercury (muscle tissue)
Selenium
Heptachlor epoxide
Hexachlorobenzene
Lindane
0
Mi rex
Toxaphene
PAH (Benzo[a]pyrene) 0.0016-0.0032
Total PCBs 0.023-0.047
24
0
0
4
0
0
0
0
0
0
0
0
2
11
0
18
0
0
8
<1
0
0
0
0
0
0
<1
3
19
Coverage: Estuarine waters of the contiguous 48 states.
Concentrations were measured in whole fish tissue. Mercury data were adjusted to
reflect concentrations in edible fillets, where mercury accumulates (adjustment
factor of 3.0, based on the available literature). All other contaminants are
presented as whole-body concentrations.
Concentrations are compared with risk guidelines for recreational fishers for four
8-ounce meals per month (U.S. EPA, 2000, 2004). Guidelines presented here are
for non-cancer risk, except for PAH, which is a cancer risk guideline.
Anorganic arsenic estimated at 2% of total arsenic.
Data source: U.S. EPA, 2005a
65
-------
INDICATOR
Coastal Fish Tissue Contaminants
References
Canadian Council of Ministers of the Environment. 1999.
Protocol for the derivation of Canadian tissue residue
guidelines for the protection of wildlife that consume
aquatic biota. Prepared by the Task Force on Water
Quality Guidelines.
Kannan, K., R.G. Smith, R.F. Lee, H.L. Wmdom, P.T.
Heimuller, J.M. Macauley, andJ.K. Summers. 1998. Dis-
tribution of total mercury and methyl mercury in water,
sediment and fish from South Florida estuaries. Arch.
Environ. Con. Tox. 34:109-118.
Mikac, N., M. Picer, P. Stegnar, and M. Tusek-Nidari.
1985. Mercury distribution in a polluted marine area, ratio
of total mercury, methyl mercury and selenium in sedi-
ments, mussels and fish. Water Res. 19:1387-1392.
Schmidt, C.J., and W.G. Brumbaugh. 1990. National
contaminant biomonitoring program: Concentrations of
arsenic, cadmium, copper, lead, mercury, selenium, and
zinc in U.S. freshwater fish 1976-1984. Arch. Environ.
Con. Toxicol. 19:731-747.
U.S. EPA (United States Environmental Protec-
tion Agency). 2005a. Data provided to ERG (an EPA
contractor) by Kevin Summers, EPA. September 2005.
U.S. EPA. 2005b. EMAP national coastal database.
Accessed 2005.
U.S. EPA. 2004. National coastal condition report II.
EPA/620/R-03/002.
U.S. EPA. 2000. Guidance for assessing chemical contami-
nant data for use in fish advisories. EPA/823/B-00/008.
Wmdom, H.L., and D.R. Kendall. 1979. Accumulation
and biotransformation of mercury in coastal and marine
biota. In: Nriagu, J.O., ed. Biogeochemistry of mercury in
the environment. Amsterdam, The Netherlands: Elsevier.
pp. 303-323.
66
-------
INDICATO
Population Served by Community Water Systems with No
Reported Violations of Health-Based Standards
Community water systems (CWS), public water systems
that supply water to the same population year-round,
served over 286 million Americans in fiscal year (FY) 2007
(U.S. EPA, 2007)—roughly 95 percent of the U.S. popula-
tion (U.S. Census Bureau, 2007). This indicator presents
the percentage of Americans served by CWS for which
states reported no violations of EPA health-based standards
for over 90 contaminants (U.S. EPA, 2004b).
Health-based standards include Maximum Contaminant
Levels (MCLs) and Treatment Techniques (TTs). An MCL is
the highest level of a contaminant that is allowed in drink-
ing water. A TT is a required treatment process (such as
filtration or disinfection) intended to prevent the occurrence
of a contaminant in drinking water (U.S. EPA, 2004c). TTs
are adopted where it is not economically or technologi-
cally feasible to ascertain the level of a contaminant, such as
microbes, where even single organisms that occur unpre-
dictably or episodically can cause adverse health effects.
Compliance with TTs may require finished water sampling,
Exhibit 3-35. U.S. population served by community water
systems with no reported violations of EPA health-based
standards, fiscal years 1993-20073
along with quantitative or descriptive measurements of
process performance to gauge the efficacy of the treatment
process. MCL-regulated contaminants tend to have long-
term rather than acute health effects, and concentrations
vary seasonally (if at all; e.g., levels of naturally occurring
chemical contaminants or radionuclides in ground water are
relatively constant). Thus, compliance is based on averages of
seasonal, annual, or less frequent sampling.
This indicator tracks the population served by CWS for
•which no violations were reported to EPA for the period
from FY 1993 to FY 2007, the latest year for which data
are available. Results are reported as a percentage of the
overall population served by CWS, both nationally and
by EPA Region. This indicator also reports the number
of persons served by systems with reported violations of
standards covering surface water treatment, microbial
contaminants (microorganisms that can cause disease), and
disinfection byproducts (chemicals that may form when
disinfectants, such as chlorine, react with naturally occur-
ring materials in water and may pose health
risks) (U.S. EPA, 2004b). The indicator is based
on violations reported quarterly by states, EPA,
and the Navajo Nation Indian Tribe, who each
review monitoring results for the CWS that
they oversee.
100
90
80
70
60
50
1 20
Q_
10
0
'93 '94 '95 '96 '97
'99 '00 '01 '02 '03 '04 '05 '06 '07
Fiscal year
Reported violations:
B New standards
(post-12/31/01) only
None
Coverage: U.S. residents served by community
water systems (CWS) (approximately 95% of
the total U.S. population).
bSeveral new standards went into effect after
12/31/01, including the Interim Enhanced
Surface Water Treatment Rule (CWS with
surface water sources serving 10,000 or more
people) and the Disinfection Byproducts (DBP) Rule for CWS that disinfect. In FY
2003, the DBP rule applied to systems serving >10,000 people; as of January 2004,
it applied to all CWS. For FY 2002-2007, each column is divided into two segments:
the lower portion reflects all standards in place at the time, while the upper portion
covers sytems with reported violations of new standards but not pre-12/31/01
standards. Adding both segments together, the total height of each column indicates
what percent of CWS customers would have been served by CWS with no reported
violations if the new standards had not gone into effect.
Data source: U.S. EPA, 2007
What the Data Show
Of the population served by CWS nationally,
the percentage served by systems for which no
health-based violations were reported for the
entire year increased overall from 79 percent
in 1993 to 92 percent in FY 2007, with a peak
of 94 percent in FY 2002 (Exhibit 3-35). This
indicator is based on reported violations of the
standards in effect in any given year. Several
new standards went into effect after December
31, 2001. These were the first new drinking
•water standards to take effect during the period
of record (beginning in 1993). The results after
FY 2001 \vould have been somewhat higher
had it not been for violations of standards that
became effective in FY 2002 or after
(Exhibit 3-35; see the dark segment atop the
columns starting in FY 2002). As EPA adds
to or strengthens its requirements for water
systems over time, compliance with standards
comes to represent a higher level of public
health protection.
When results are broken down by EPA
Region, some variability over time is evi-
dent (Exhibit 3-36). Between FY 1993 and
FY 2007, most Regions \vere consistently
67
-------
INDICATO
Population Served by Community Water Systems with No
Reported Violations of Health-Based Standards
Exhibit 3-36. U.S. population served by
community water systems with no reported
violations of EPA health-based standards, by
EPA Region, fiscal years 1993-2007ab
100
90
80
70
60
50
40
30
20
10
0
-R2
-R3
R4
-R5
R6
R7
R8
-R9
-R10
-Nat'l
'93 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '05 '06 '07
Fiscal year
Coverage: U.S. residents served
by community water systems
(CWS) (approximately 95% of
the total U.S. population).
bBased on reported violations of
the standards in effect in any
given year.
Data source: U.S. EPA, 2007
above the national percentage. Three of the Regions were
substantially below the national average over much of
the period of record, but as of FY 2007, only one Region
remained well below the national percentage, largely
because of a small number of public water systems serving
large populations.
In FY 2007, reported violations involving surface
•water treatment rules in large CWS were responsible for
exceeding health-based standards for 8.9 million people
(3.1 percent of the population served by CWS nation-
ally) (Exhibit 3-37). Reported violations of heath-based
coliform standards affected 10.6 million people (3.7 percent
of the CWS-served population), and reported violations of
the health-based disinfection byproducts standards (Stage
1) affected 3.6 million people (1.3 percent of the CWS-
served population). Overall, of the 8.5 percent of the
population served by systems with reported violations in
FY 2007, 84 percent of these cases involved at least one of
these three rules governing treatment to prevent -water-
borne diseases—the most -widespread and acute threat to
health from drinking -water—or the contaminants created
by such treatment.
Exhibit 3-37. U.S. population served by
community water systems with reported
violations of EPA health-based standards, by
type of violation, fiscal year2007a
Percent of
Population served CWS customers
Any violation
Selected violations
Stage 1 Disinfection
Byproducts Rule
Surface Water
Treatment Rules
Total Coliform Rule
Any of these
selected rules
24,279,892
3,643,104
8,945,673
10,569,935
20,472,902
8.5
1.3
3.1
3.7
7.1
Coverage: U.S. residents served by community water systems
(CWS) (approximately 95% of the total U.S. population).
bSome CWS violated more than one of the selected rules.
Data source: U.S. EPA, 2007
Indicator Limitations
• Non-community -water systems (typically relatively small
systems) that serve only transient populations such as
restaurants or campgrounds, or serving those in a non-
domestic setting for only part of their day (e.g., a school,
religious facility, or office building), are not included in
population served figures.
• Domestic (home) use of drinking -water supplied by pri-
vate \vells—-which serve approximately 15 percent of the
U.S. population (USGS, 2004)—is not included.
• Bottled -water, -which is regulated by standards set by the
Food and Drug Administration, is not included.
• National statistics based on population served can be
volatile, because a single very large system can sway
the results by up to 2 to 3 percent; this effect becomes
more pronounced -when statistics are broken down at the
regional level, and still more so for a single rule.
• Some factors may lead to overstating the extent of
population receiving -water that violates standards. For
example, the entire population served by each system in
68
-------
INDICATO
Population Served by Community Water Systems with No
Reported Violations of Health-Based Standards
violation is reported, even though only part of the total
population served may actually receive water that is out
of compliance. In addition, violations stated on an annual
basis may suggest a longer duration of violation than may
be the case, as some violations may be as brief as an hour
or a day.
• Other factors may lead to understating the popula-
tion receiving water that violates standards. CWS that
purchase water from other CWS are not always required
to sample for all contaminants themselves, and the CWS
that are -wholesale sellers of water generally do not report
violations for the population served by the systems that
purchase the water.
• Under-reporting and late reporting of violations by
states to EPA affect the ability to accurately report the
national violations total. For example, EPA estimated
that between 1999 and 2001, states were not reporting
35 percent of all health-based violations, which reflects a
sharp improvement in the quality of violations data com-
pared to the previous 3-year period (U.S. EPA, 2004a).
• State data verification and other quality assurance analy-
ses indicate that the most -widespread data quality prob-
lem is under-reporting of monitoring and health-based
violations and inventory characteristics. Under-reporting
occurs most frequently in monitoring violations; even
though these are separate from the health-based viola-
tions covered by the indicator, failures to monitor could
mask violations of TTs and MCLs.
Data Sources
Data for this indicator -were obtained from EPA's Safe
Drinking Water Information System (U.S. EPA, 2007)
(http://www.epa.gov/safewater/data/getdata.html;
http://www.epa.gov/safewater/data/pivottables.html). This
database contains a record of violations reported to EPA by
the states or other entities that oversee CWS, along -with
annual summary statistics.
References
U.S. Census Bureau. 2007. Monthly Population Estimates
for the United States: April 1, 2000 to September 1, 2007.
NA-EST2006-01. Accessed October 2007. .
Available from
U.S. EPA (United States Environmental Protection
Agency). 2007. Safe Drinking Water Information System,
Federal Version. Accessed October 2007.
U.S. EPA. 2004a. Safe Drinking Water Act 30th anni-
versary fact sheet: Drinking -water monitoring, compli-
ance, and enforcement,
U.S. EPA. 2004b. Safe Drinking Water Act 30th anniver-
sary fact sheet: Drinking -water standards and health effects.
U.S. EPA. 2004c. Safe Drinking Water Act 30th anniver-
sary fact sheet: Glossary,
USGS (United States Geological Survey). 2004. Estimated
use of-water in the United States in 2000. 2004 revision.
69
-------
Land Cover
Land cover represents the actual or physical presence
of vegetation (or other materials where vegetation is
nonexistent) on the land surface. Land cover is also often
described as what can be seen on land when viewed from
above. Land cover is one means to categorize landscape
patterns and characteristics, and is critical in understanding
the condition of the environment, including the availabil-
ity of habitat, changes in habitat, and dispersion and effects
of chemicals and other pollutants in and on the environ-
ment. For the purposes of this indicator, land cover is
described in terms of six major classes: forest, grass, shrub,
developed, agriculture, and other (includes ice/snow, bar-
ren areas, and wetlands). A seventh category, water, is not
discussed as a land cover type in this chapter. See Chapter
3 for more information on trends related to water. More
information about forest land can be found in the Forest
Extent and Type indicator and wetland acreage is discussed
in greater detail in the Wetlands indicator.
In 1992, several federal agencies agreed to operate as a
consortium, known as the Multi-Resolution Land Char-
acteristics (MRLC) Consortium, to acquire and analyze
satellite-based remotely sensed data for environmental
monitoring programs (MRLC Consortium, 2006). The
initial result of the MRLC effort was development of the
1992 National Land Cover Dataset (NLCD), which, until
recently, was the only comprehensive recent classification of
land cover in the contiguous U.S. (USGS, 2007). In 2007,
the MRLC Consortium published the 2001 National Land
Cover Database, an updated and improved version of the
1992 NLCD (Homer et al, 2007). The database provides
information about 16 land cover classes at a 30-meter reso-
lution, comprising approximately 27 billion cells covering
Exhibit 4-1. Land cover of the contiguous U.S., based on 2001 NLCD£
H- ^*]W''
P\ V-&JB
••>v£jJi*. 'i
•r
aSee box in text for definitions of land
cover categories.
Data source: U.S. EPA, 2007b
Agriculture Developed Forest cover
Cultivated crops • High-density (impervious£80%) • Deciduous forest
Pasture/hay • Medium-density (impervious 50-79%) • Evergreen forest
• Low-density (impervious 20-49%) • Mixed forest
• Open space (impervious <20%)
Grass cover
Grassland
Shrub cover
Shrubland
Other
D Perennial ice/snow
Barren
• Woody wetland
• Emergent herbaceous wetland
Water
• Open water
70
-------
Land Cover
Exhibit 4-2. Land cover types in the U.S., based
on 2001 NLCD and FIAabc
Agriculture
448.9 million acres
(21.9%)
Grass cover
290.5 million
acres
(14.1%)
Shrub cover
419.2 million acres
(20.4%)
Otherd
117.7 million
acres
(5.7%)
Water
33.5 million Developed
acres 102.5 million acres
(1.6%) (5.0%)
Coverage: All surface area of the contiguous 48 states, plus forest
land in Alaska and Hawaii.
bSee box in text for definitions of land cover categories.
cTotals may not add to 100% due to rounding.
d"0ther" includes ice/snow, barren areas, and wetlands.
Data source: Smith etal., 2004; U.S. EPA, 2007b
the contiguous U.S., based on Landsat images from 1999
to 2002. Due to differences in methodology, direct com-
parison of the 1992 and 2001 NLCD data sets does not
currently provide valid trend data. Efforts are underway to
develop an algorithm that will allow such comparisons in
the near future.
This indicator represents data from the 2001 NLCD and
the U.S. Department of Agriculture Forest Service's Forest
Inventory and Analysis (FIA), which uses a statistical survey
design and comparable methods to assess the extent, type,
age, and health of forests on private and public land in all
states. The 2001 NLCD provides a synoptic classification of
land cover, but does not include Alaska and Hawaii, thereby
classifying only 1.92 billion acres out of approximately 2.3
billion acres of land in the U.S. To supplement the NLCD,
data from the 2001 FIA were used to provide forest cover
estimates in Alaska and Hawaii (128.6 million acres).
For this indicator, the 16 land cover classes created in the
NLCD were aggregated into the six major land cover types
described above, along with water (Heinz Center, 2005).
What the Data Show
The combination of the NLCD for the contiguous 48
states and the FIA for forest cover estimates in Alaska and
Exhibit 4-3. Land cover types in the U.S. by
EPA Region, based on 2001 NLCD and FIAat
400
350
_ 300
CO
CD
& 250
c
:§ 200
— 150
ro
CD
< 100
50
D Other0
• Developed
D Shrub cover
D Grass cover
D Agriculture
• Forest cover
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10
EPA Region
Coverage: All land area of the
contiguous 48 states (excluding
water), plus forest land in Alaska
and Hawaii.
bSee box in text for definitions of
land cover categories.
c"0ther" includes ice/snow, barren
areas, and wetlands.
Data source: Smith et a/., 2004; U.S. EPA, 2007b
EPA Regions
Hawaii shows approximately 641 million acres of forest,
449 million acres of agriculture, 419 million acres of shrub,
291 million acres of grass, and 103 million acres of devel-
oped cover types (Exhibits 4-1 and 4-2).
NLCD and FIA data show variation in cover types by
EPA Region, with forest dominating in Regions 1, 2, 3, 4,
and 10; agriculture in Regions 5 and 7; grass in Region 8;
and shrub in Region 6 and 9 (Exhibit 4-3). Two-thirds of
the grass acreage in the nation is located in Regions 6 and
8, nearly two-thirds of shrub acreage is in Regions 6 and
9, and nearly half the forest acreage is in Regions 4 and 10
(including Alaska).
Indicator Limitations
• Trend data are not available for this indicator. Land
cover data for the entire nation at adequate resolution
to support this indicator are currently available for two
points in time (1992 and 2001). However, due to differ-
ences in methodology in creation of the data sets, they
are not directly comparable. The MRLC Consortium
is developing a change product intended to enable valid
comparisons of the two data sets (MRLC Consortium,
2007a,b). The product is scheduled to be available in 2008.
Until this project is completed, there are no consistent,
comprehensive, nation-wide data to describe trends in land
cover at the national or EPA Regional levels.
71
-------
INDICATOR
Land Cover
Definitions of Land Cover Categories for Exhibits 4-1, 4-2, and 4-3
Agricultural (NLCD 2001 definition): Areas charac-
terized by herbaceous vegetation that has been planted;
is intensively managed for the production of food, feed,
or fiber; or is maintained in developed settings for spe-
cific purposes. Herbaceous vegetation must account for
75 to 100 percent of the cover. Includes the "orchards/
vineyards/other" subcategory, which covers areas
planted or maintained for the production of fruits, nuts,
berries, or ornamentals. Includes two subcategories:
"pasture/hay" and "cultivated crops."
Developed (NLCD 2001 definition): Areas charac-
terized by a high percentage (30 percent or greater) of
constructed materials (e.g., asphalt, concrete, buildings).
Includes four subcategories: "Developed, open space"
(less than 20 percent impervious surface), "Developed,
low intensity" (20-49 percent impervious surface),
"Developed, medium intensity" (50-79 percent imper-
vious surface), and "Developed, high intensity" (80
percent or more impervious surface).
Shrubland (NLCD 2001 definition): Areas charac-
terized by natural or semi-natural woody vegetation
•with aerial stems, generally less than 6 meters tall, with
individuals or clumps not touching or interlocking.
Both evergreen and deciduous species of true shrubs,
young trees, and trees or shrubs that are small or stunted
because of environmental conditions are included.
Grassland (NLCD 2001 definition): Upland areas
dominated by grammanoid or herbaceous vegetation,
generally greater than 80 percent of the total vegetation.
These areas are not subject to intensive management,
such as tilling, but can be utilized for grazing.
FIA data for forest land in Alaska and Hawaii were used
to complement the NLCD because NLCD data do not
currently exist for these states, although they are planned
for late 2007. Ongoing data collection under both the
FIA and the NLCD is needed to assess land cover trends.
National estimates of land cover vary, depending on the
survey approach, data sources, classification, timing, etc.
The interaction of these variables will result in different
estimates of the extent of any given land cover category
depending on the data set used. Techniques relying on
satellite data to generate land cover estimates classify what
is visible from above, meaning they may underestimate
developed cover in heavily treed urban areas and underes-
timate forest cover where trees have been harvested. For
example, National Resources Inventory (USDA NRCS,
2007) estimates for developed land are 6 percent above
Forest (NLCD 2001 definition): Areas characterized
by tree cover (natural or semi-natural woody vegeta-
tion, generally greater than 6 meters tall); tree canopy
accounts for 25 to 100 percent of the cover.
Forest (FIA definition): Land at least 10 percent
stocked by forest trees of any size, including land that
formerly had such tree cover and that will be naturally
or artificially regenerated. Forest land includes transi-
tion zones, such as areas between heavily forested and
nonforested lands that are at least 10 percent stocked
•with forest trees and forest areas adjacent to urban and
built-up lands. Also included are pinyon-juniper and
chaparral areas in the West and afforested areas. The
minimum area for classification of forest land is 1 acre.
Roadside, streamside, and shelterbelt strips of trees must
have a crown width of at least 120 feet to qualify as
forest land. Unimproved roads and trails, streams, and
clearings in forest areas are classified as forest if less than
120 feet \vide. (FIA data are used in Alaska and Hawaii,
due to lack of NLCD availability.)
Other: Includes NLCD 2001 snow, ice, wetlands, and
barren. Barren areas are defined as areas of bedrock, des-
ert pavement, scarps, talus, slides, volcanic material, gla-
cial debris, sand dunes, strip mines, gravel pits, and other
accumulations of earthen material. Generally, vegetation
accounts for less than 15 percent of total cover.
Sources: U.S. EPA, 2007a; Smith et al., 2004.
the NLCD estimates and FIA estimates of forestland in
2002 are nearly 17 percent above the NLCD.
No standardized land cover classification system is cur-
rently used among federal agencies. As a result of this
limitation, there is no consistency in the assessment of
land cover trends across agencies.
Data Sources
Land cover data for the contiguous 48 states were obtained
from the NLCD (U.S. EPA, 2007b). These data were
grouped into the major land cover categories as described
by the Heinz Center (2005) (see technical note for the
Heinz Center's "Ecosystem Extent" indicator). Forest cover
estimates for 2002 in Alaska and Hawaii were obtained
from a report published by the FIA program (Smith et al.,
2004). FIA data in this report have a nominal date of 2002
72
-------
Land Cover
but represent the best data available at the end of the 2001
field season for each state.
References
Heinz Center (The H. John Heinz III Center for Sci-
ence, Economics, and the Environment). 2005. The state
of the nation's ecosystems: Measuring the lands, waters,
and living resources of the United States. New York, NY:
Cambridge University Press. Web update 2005:
Homer, C., J. Dewitz, J. Fry, M. Coan, N. Hossain, C.
Larson, N. Herold, A. McKerrow, J.N. VanDriel, andj.
Wickham. 2007. Completion of the 2001 National Land
Cover Database for the conterminous United States.
Photogramm. Eng. Rem. S. 73(4):337-341.
MRLC Consortium. 2007a. Comparison of NLCD 1992
and NLCD 2001. Accessed December 2007.
MRLC Consortium. 2007b. Frequently asked questions.
Accessed December 2007.
MRLC Consortium. 2006. About the MRLC program.
Accessed January 2006.
Smith, W.B., P.O. Miles, J.S. Vissage, and S.A. Pugh.
' J o ' o
2004. Forest resources of the United States, 2002. USDA
Forest Service.
USDA NRCS (United States Department of Agriculture,
Natural Resources Conservation Service). 2007. National
Resources Inventory, 2003 annual NRI: Land use.
U.S. EPA (United States Environmental Protection
Agency). 2007a. NLCD classification schemes. Accessed
November 28, 2007.
U.S. EPA. 2007b. NLCD 2001 data provided to ERG (an
EPA contractor) by James Wickham, Office of Research
and Development. November 29, 2007.
USGS (United States Geological Survey). 2007. National
Land Cover Dataset 1992 (NLCD 1992). Accessed
November 2007.
73
-------
Land Cover in the Puget Sound/Georgia Basin
British
Columbia/1'
Washington
Changes in land use and corresponding changes in land
cover can alter the basic functioning and resilience of
ecological systems. Watersheds, for example, experience
a cascade of effects among critical physical, chemical, and
biological processes when land cover changes (NWP, 1995;
Thorn and Borde, 1998). For instance, removal of vegeta-
tion can increase erosion, leading to impacts on soil and
•water quality, and increases in developed land typically
result in a corresponding increase in imper-
vious surfaces with consequences for runoff,
among other issues. While individual
impacts to a landscape may appear as small
changes, the combined impacts of particu-
lar land uses or land management practices
on -watersheds can have substantial effects
on \vater quality, species composition, and
flooding patterns (PSAT, 2002, 2004). Such
combined impacts are often referred to as
"cumulative effects." As a result of their
potential to broadly and substantially influ-
ence environmental condition, land cover
and use are important factors to monitor.
This indicator compares changes in two
land cover metrics for the Puget Sound and
Georgia Basin in Washington state and part
of British Columbia, Canada. The metrics
include percent change of urban and forest
land cover. Data cover the period from 1995
to 2000 for the U.S. portion of the basin and
from 1992 to 1999 for the Canadian side of
the basin. The metrics represent the change
in total urban or forested land area divided
by total land area in the -watershed. Forest
and urban land cover are two of the most
important factors affecting the condition of
•watersheds in the Puget Sound Basin (Alberti
and Marzluff, 2004; Alberti, 2005). In con-
trast to the nation-wide land cover indicator,
which is based on NLCD data, this indicator
relies on data derived from four assembled
USGS Landsat scenes covering the U.S. por-
tion of the Puget Sound Basin and from a
combined scene covering the Canadian land
area. The land cover data for all USGS 6th
field -watersheds in the basin -were produced
from NOAA Coastal Change Analysis
Program (C-CAP) data and from Canadian
Baseline Thematic Mapping (BTM) data.
The USGS Hydrologic Unit Codes and
Canadian -watershed groupings provide topo-
graphically delineated -watersheds, -which are
aggregated, or "nested," into larger sub-basin
and basin units.
What the Data Show
Forest Cover
Little or no change in forest cover -was observed in 2,068
•watersheds (76 percent) of the 2,725 -watersheds assessed
(Exhibit 4-4, panel A). However, 279 -watersheds (10 per-
cent) saw at least 2.5 percent of their mature forest cover
converted to some other land cover, often bare ground,
Exhibit 4-4. Land cover change in watersheds of the
Puget Sound/Georgia Basin, 1992-2000ab
A. Forest cover
B. Urbanization
Percent change in forest cover:
• -5.0% or more n No change
• -4.9% to-2.5% 0.1% to 0.85%
-2.4% to-0.1% More than 0.85%
British
Columbia
Washington
Percent change in urbanization:
n No change • 0.301 % to 0.70%
0.05% or less • 0.71 % to 1.93%
0.051 % to 0.30% • More than 1.93%
2,500
2,000
B 1,500
^ 1,000
2,068
2,500
i 2,000
S 1,500
* 1,000
2.329
Percent change
Percent change
Coverage: 2,725 watersheds within the Puget Sound/Georgia Basin, located in the
state of Washington and the Canadian province of British Columbia. U.S. watersheds
are 12-digit Hydrologic Unit Code (HUC12) watersheds.
'U.S. data reflect changes from 1995 to 2000, while Canadian data reflect changes from
1992to1999.
Data source: British Columbia Integrated Land Management Bureau, 2001; CommEn
Space, 2005; NOAA, 2006
74
-------
INDICATOR
Land Cover in the Puget Sound/Georgia Basin
immature vegetation, or industrial/urban uses. At the same
time, another group of 205 -watersheds (8 percent), gener-
ally those at higher elevations, indicated a net increase in
forest cover as young stands or cleared areas have re-grown
into more mature forest cover classes.
Urbanization
During the same period, little or no change in urban land
cover was observed in approximately 90 percent of the
2,725 assessed -watersheds -within the basin (Exhibit 4-4,
panel B). However, urbanization increased across many
low-elevation -watersheds and shoreline areas, -with 158
•watersheds (6 percent) expanding the urban portion of the
•watershed by between 0.7 and 1.93 percent, and another 58
•watersheds (2 percent) showing increases of more than 1.93
percent. Research has shown that as a -watershed's drain-
age area becomes paved or other-wise impervious, there is a
high potential for physical, chemical, and biological impair-
ments to both -water quality conditions and other aquatic
resources (NWP, 1995; Alberti and Marzluff, 2004).
Indicator Limitations
• While the U.S. C-CAP data and the Canadian BTM
data have similar and overlapping time periods, as cur-
rently presented, the U.S. data reflect change from 1995
to 2000 and the Canadian data reflect change from 1992
to 1999.
• The size of the data pixels and the minimum mapping
unit size affect the classification of certain features such as
narrow riparian corridors, and can affect the percentages
in the indicators.
Data Sources
The full analysis has not been published as a data set,
but it is based on publicly available data sets compiled by
CommEn Space (http://www.commenspace.org). Raw
data for the U.S. portion of this indicator are available
from C-CAP (NOAA, 2006), and Canadian data are
available from the British Columbia Integrated Land
Management Bureau (2001). Additional technical back-
ground is provided by U.S. EPA (2006).
References
Alberti, M. 2005. The effects of urban patterns on
ecosystem function. Int. Regional Sci. Rev. 28(2):168-192.
Alberti, M., andj. Marzluff 2004. Resilience in urban
ecosystems: Linking urban patterns to human and ecologi-
cal functions. Urb. Ecosyst. 7:241-265.
British Columbia Integrated Land Management Bureau.
2001. Baseline thematic mapping,
CommEn Space. 2005. Cartography services provided
to EPA.
NOAA (National Oceanic and Atmospheric Administra-
tion). 2006. Coastal Change Analysis Program (C-CAP)
database. Accessed 2007.
NWP (North-west Forest Plan). 1995. Ecosystem analysis at
the -watershed scale: Federal guide for -watershed analysis.
Portland, OR: USFS Regional Ecosystem Office.
PSAT (Puget Sound Action Team). 2004. State of the
Sound. Report to the Washington state legislature.
Olympia, WA.
PSAT. 2002. Puget Sound update. Eighth report of the
Puget Sound ambient monitoring program. Olympia, WA.
Thorn, R., and A. Borde. 1998. Human intervention in
Pacific North-west coastal ecosystems. In: McMurray,
G.R., and RJ. Bailey, eds. Change in Pacific North-west
coastal ecosystems. NOAA Coastal Ocean Program Deci-
sion Analysis Series No. 11.
U.S. EPA (United States Environmental Protection
Agency). 2006. Puget Sound Georgia Basin ecosystem
indicators: Ecosystem indicator references and technical
background.
75
-------
Land Use
Land use is the purpose of human activity on the land.
Unlike land cover, land use may not always be vis-
ible. For example, a unit of land designated for use as
timberland may appear identical to an adjacent unit of
protected forestland or, if recently harvested, may appear
not to be in forest land cover at all. Land use is generally
designated through zoning or regulation and is one of the
most obvious effects of human inhabitation of the planet.
It can affect both human health and ecological systems,
for example by changing the hydrologic characteristics of
a -watershed, the potential of land to erode, the condition
or contiguity of plant and animal habitat, or the spread of
vector-borne diseases.
This indicator tracks trends in acreages of major land
uses over the 1977-2003 period using several data sources.
These sources do not always cover the same time period,
sample the same resource or geography, or use the same
definitions, but each of them provides an important piece
of the land use picture over time. Definitions for the vari-
ous land use categories in this indicator can be found on
the following page.
The National Resources Inventory (NRI) conducted
by the U.S. Department of Agriculture (USDA) Natural
Resources Conservation Service was used to track trends
in "crop and pasture" land (row crop, orchard, and pasture
uses) and "developed" land (residential, commercial, indus-
trial, and transportation uses). The NRI developed esti-
mates every 5 years on non-federal lands in the contiguous
U.S. between 1977 and 1997, and annual estimates based
on a smaller sample size beginning in 2001.
The Forest Inventory and Analysis (FIA) surveys con-
ducted by the USDA Forest Service were used to track
trends in forest and timberlands. The FIA surveys include
both private and public land in all 50 states. The FIA previ-
ously assessed forest and timberland acreage every 10 years,
but the data are now updated on a rolling basis using sur-
veys that sample a different portion of FIA sites every year.
The USDA National Agricultural Statistics Service
(NASS) Census of Agriculture was used to track trends in
the extent of cropland, cropland used only for pasture, pas-
tureland, and rangeland. NASS data are available for 1997
and 2002 only. Data on the extent of grass and forested
Exhibit 4-5. Land use trends in the U.S., 1977-20033
800
700
600
500
400
300
200
100
0
NRI D Pasture
D Cropland
131.1
419.9
125.2
119.5
1170
NASS D Cropland pasture
D Cropland
NASS D Pastureland and rangeland
381.3
376.4
367.9
66.4
378.9
373.6
398.2
395.3
1977 1982 1987 1992 1997 2003 1977 1982 1987 1992 1997 2002 1977 1982 1987 1992 1997 2002
ERS [
[
H Forested grazing land
H Grassland pasture and range
158
597
155
591
145
591
141
580
134
587
FIA D Timberland
492.4
486.3
503.7
503.5
NRI D Developed
|72.9||79.5||86.5||98.3 108.1
800
700
600
500
400
300
200
100
° 1977 1982 1987 1992 1997 2002 1977 1982 1987 1992 1997 2002 1977 1982 1987 1992 1997 2003
Year
aSee box in text for definitions of land use categories.
Data source: Lubowski et al., 2006; Smith et al., 2004; USDA NASS, 2004; USDA NRCS, 2007
76
-------
Land Use
Exhibit 4-6. Land use in the U.S. by EPA Region, 2002-20033
250
200
150
100
50
FIA
Timberland
(2002)
NASS
Cropland
(2002)
ERS
Rangeland
(2002)
NRI
Developed
(2003)
R1
R2 R3
R4
R5 R6
EPA Region
R7
R9 R10
aSee box in text for definitions of land use categories.
Data source: Lubowski et al., 2006; Smith et al., 2004;
USDA NASS, 2004; USDA NRCS, 2007
EPA Regions
rangeland (typically "unimproved" grazing land) are avail-
able from the USDA Economic Research Service (ERS) for
5-year intervals from 1982 through 2002.
What the Data Show
The acreage of lands used for growing food and forage
crops has declined since 1982, while developed land has
increased and timberland has remained approximately con-
stant (Exhibit 4-5). As of 2002-2003, estimates from both
the NRI (2003 data) and the NASS (2002 data) indicate
that between 368 and 374 million acres were used for food
crop production, approximately 16 percent of the U.S. land
area. Estimates of pasture or land used to support forage
for livestock vary, depending on the definitions. The NRI
classifies 117 million acres as pasture, while the NASS clas-
sifies about 61 million acres as cropland used for pasture.
The NASS classifies more than 395 million additional
acres as pasture or rangeland for grazing. The broader
ERS estimate of land available for grazing totals about
587 million acres, and includes grassland and other non-
forested pasture and range. If forest lands used for grazing
are also included, the total ERS estimate for these lands is
721 million acres for 2002. The NASS shows a decrease in
the extent of cropland (5 million acres), cropland pasture
(6 million acres), and pastureland and rangeland (3 million
acres) between 1997 and 2002. The NRI data suggest that
these declines are part of a longer trend, with NRI crop-
land and pasture declining by slightly more than 66 million
acres (12 percent) between 1982 and 2003.
ERS data also show a downward trend
for pasture and rangeland between 1982
and 2002, with the largest decrease being
a 24-million-acre (15 percent) decline in
forest land used for grazing. According to
the NRI, 5 percent (108.1 million acres) of
U.S. land area was considered developed1
as of 2003 (Exhibit 4-5). This represents a
gain of 48 percent (35.2 million acres) since
1982. While the amount of developed land
is a small fraction of the total, its ecologi-
cal impact can be disproportionately high
relative to other land use types. Paving and
the creation of other impervious surfaces
can change local hydrology, climate, and
carbon cycling, leading to increased surface
runoff, pollution, and degradation of-wet-
lands and riparian zones.
Forest lands are managed by a complex
array of interests to meet multiple purposes,
including providing habitat for a variety of
species, recreation, and timber production.
While forest is a land cover classification,
timberland is a land use classification that
reflects forest land capable of producing
at least 20 cubic feet per acre per year of industrial wood
and not withdrawn from timber utilization by statute or
regulation. Approximately 504 million acres of U.S. forest
land, or 22 percent of the total U.S. land area, qualified as
timberland in 2002 (Exhibit 4-5). This total reflects a net
gain of about 11 million acres (2 percent) between 1977
and 2002, which the FIA attributes largely to reversion
of abandoned lands and reclassification of some National
Forest lands to align with classifications used on other land
ownerships (Smith et al., 2004).
Land use varies widely by EPA Region (Exhibit 4-6).
According to the most recent data for each land use type,
Regions 6, 8, and 9 together have more than three-
quarters of the nation's grazing land, while Region 4
has the largest portion of timberland (27 percent of total
U.S. timberland). Trends also vary widely among regions.
About 83 percent of the cropland lost between 1987 and
2003 was in five EPA Regions (Regions 4, 5, 6, 7, and
8) (Exhibit 4-7, panel A). Increases in developed land are
responsible for part of this decline; for example, developed
land increased by nearly 60 percent from 1987 to 2003 in
Region 4 (Exhibit 4-7, panel B). Other factors include the
federal Conservation Reserve Program, which has assisted
private landowners in converting about 35 million acres of
The land use classification for developed land uses NRI data and is con-
siderably different from the land cover classification for developed land,
which uses NLCD data. See Section 4.2 for more information.
77
-------
Land Use
Exhibi
10
_ 8
CO
ro
i 4
Change in area (m
30 oacn-krborv
60
50
8 40
ro 30
c
'§> 20
J 10
o
= 0
CD
1 -10
Q_
-20
-30
-40
aSee bo>
Datas
14-7.
Changes in land use in the U.S. by EPA Region, 1977-20033
A. Acreage change
NRI developed
(1987-2003)
. 1.
NRI crop
& pasture
(1987-2003)
~n
'
NASS crop
& pasture
(1997-2002)
MM " "
FIATimberland
(1977-1987) (1987-1997)
1
*
1
(1997-2001)
B. Percent change
NRI
(19
developed
87-2003)
NRI crop
& pasture
(1987-2003)
UN' »
NASS crop
& pasture
(1997-2002)
|-H
FIATimberland
(1977-1987) (1987-1997)
(1997-2001)
_• I
< in text for definitions of land use categories.
jurce: Smith etal., 2004; USDA NASS, 2004; USDA NRCS, 2007
• R1
• R2
R3
R4
• R5
R6
R7
R8
• R9
R10
EPA Regions
3D
\r %
highly erodable cropland to vegetative cover since 1985 (as
of 2004) (USDA Farm Service Agency, 2004).
Indicator Limitations
• Estimates are derived from a variety of inventories and
samples, conducted over different time periods and for
different purposes. This limits the ability to integrate the
data and track changes over time.
The NRI does not report land use data for Alaska, which
encompasses 365 million acres of the 2.3 billion acres
nation-wide. The NRI also does not provide data on
federal lands (representing 20 percent of the contiguous
U.S. land and one-third of Alaska). Because federal land
is seldom used for agriculture or urban development, and
there is relatively little developed or agricultural land in
Alaska, the NRI data likely offer a reasonable approxi-
mation of national trends in these categories.
78
-------
INDICATOR
Land Use
Definitions of Land Use Categories for Exhibits 4-5. 4-6. and 4-7
NRI (USDA NRCS, 2004)
Developed: A combination of land cover/use catego-
ries: urban and built-up areas and rural transportation land.
Urban and built-up areas. A land cover/use cat-
egory consisting of residential, industrial, commer-
cial, and institutional land; construction sites; public
administrative sites; railroad yards; cemeteries; air-
ports; golf courses; sanitary landfills; sewage treatment
plants; water control structures and spillways; other
land used for such purposes; small parks (less than 10
acres) within urban and built-up areas; and high-ways,
railroads, and other transportation facilities if they are
surrounded by urban areas. Also included are tracts of
less than 10 acres that do not meet the above definition
but are completely surrounded by urban and built-up
land. Two size categories are recognized in the NRI:
areas of 0.25 acre to 10 acres, and areas of at least 10
acres.
Large urban and built-up areas. A land cover/
use category composed of developed tracts of at
least 10 acres—meeting the definition of urban and
built-up areas.
Small built-up areas. A land cover/use category
consisting of developed land units of 0.25 to 10
acres, which meet the definition of urban and
built-up areas.
Rural transportation land. A land cover/use cat-
egory \vhich consists of all high-ways, roads, railroads
and associated right-of-ways outside urban and built-
up areas; also includes private roads to farmsteads or
ranch headquarters, logging roads, and other private
roads (field lanes are not included).
Cropland: A land cover/use category that includes areas
used for the production of adapted crops for harvest.
Two subcategories of cropland are recognized: cultivated
and noncultivated. Cultivated cropland comprises land in
row crops or close-grown crops and also other cultivated
cropland, for example, hay land or pastureland that is in
a rotation -with row or close-grown crops. Noncultivated
cropland includes permanent hay land and horticultural
cropland.
Pastureland: A land cover/use category of land man-
aged primarily for the production of introduced forage
plants for livestock grazing. Pastureland cover may consist
of a single species in a pure stand, a grass mixture, or a
grass-legume mixture. Management usually consists of
cultural treatments: fertilization, -weed control, reseeding
or renovation, and control of grazing. For the NRI,
pastureland includes land that has a vegetative cover of
grasses, legumes, and/or forbs, regardless of whether or
not it is being grazed by livestock.
FIA (Smith etal., 2004)
Forest land: Land at least 10 percent stocked by forest
trees of any size, including land that formerly had such
tree cover and that -will be naturally or artificially regen-
erated. Forest land includes transition zones, such as areas
between heavily forested and nonforested lands that are
at least 10 percent stocked -with forest trees and forest
areas adjacent to urban and built-up lands. Also included
are pinyon-juniper and chaparral areas in the West and
afforested areas. The minimum area for classification of
forest land is 1 acre. Roadside, streamside, and shelter-
belt strips of trees must have a crown -width of at least
120 feet to qualify as forest land. Unimproved roads and
trails, streams, and clearings in forest areas are classified
as forest if less than 120 feet -wide.
Timberland: Forest land that is producing or can pro-
duce crops of industrial -wood and is not withdrawn from
timber utilization by statute or administrative regulation.
(Areas qualifying as timberland must be able to produce
more than 20 cubic feet per acre per year of industrial
•wood in natural stands. Currently inaccessible and inop-
erable areas are included.)
MASS (USDA NASS, 2004)
Cropland: A category including cropland harvested,
cropland idle or used for cover crops or soil improve-
ment but not harvested and not pastured, cropland
on -which all crops failed, and cropland in cultivated
summer fallow. Not included is cropland used only for
pasture or grazing.
Cropland pasture: Cropland used only for pasture or
grazing, -which could have been used for crops -without
additional improvement. Also included are acres of crops
hogged or grazed but not harvested prior to grazing.
However, cropland pastured before or after crops -were
harvested counts as harvested cropland rather than crop-
land for pasture or grazing.
Pastureland and rangeland: All grazable land—
irrigated or dry—that does not qualify as cropland or
•woodland pasture. In some areas, this is high-quality
pastureland but cannot be cropped -without improve-
ments. In others, it can barely be grazed and is only
marginally better than -waste land.
79
-------
Land Use
ERS (Lubowski et al., 2006)
Grassland pasture and range: All open land used
primarily for pasture and grazing, including shrub and
brush land types of pasture; grazing land with sagebrush
and scattered mesquite; and all tame and native grasses,
legumes, and other forage used for pasture or graz-
ing. Because of the diversity in vegetative composition,
grassland pasture and range are not always clearly distin-
guishable from other types of pasture and range. At one
extreme, permanent grassland may merge with cropland
pasture; grassland is also often found in transitional areas
•with forested grazing land.
Forested land grazed: Forested grazing land consists
mainly of forest, brush-grown pasture, arid woodlands,
and other areas within forested areas that have grass or
other forage growth. The total acreage of forested graz-
ing land includes woodland pasture in farms plus esti-
mates of forested grazing land not in farms. For many
states, the estimates include significant areas grazed only
lightly or sporadically. The Census of Agriculture, the
National Resources Inventory, and the Forest Inventory
and Analysis are the principal sources of data.
NRI data use three subcategories of types of developed
land: large built-up areas, small built-up areas, and rural
transportation land. Because ecological effects from
developed land depend on the density of development
and many other factors, the limited NRI categories are
not discriminating enough to support detailed analyses of
ecological effects of developed land.
• The FIA data are aggregated from state inventories in
many cases, and dates of data collection for these inven-
tories vary by state—for example, ranging from 1980 to
2001 for reporting 2002 estimates.
Some land uses may be administratively designated but
not physically visible (e.g., lands that are reserved for
parks or wilderness may appear similar to lands that are
managed for natural resources).
Land use designations are most frequently managed
and monitored by local governments, each using differ-
ent approaches and classifications. This makes national
summaries difficult.
• The extent of lands used for energy production, resource
extraction, or mining is not known and represents a
data gap.
Lands specifically protected for certain uses such as -wil-
derness or parks have been periodically inventoried for
the nation. These statistics are currently not reported in a
form that allows comparison with other statistics.
Data Sources
Data \vere obtained from several original sources and
compiled by EPA Region. ERS data were obtained from
Lubowski et al. (2006). FIA data \vere obtained from Smith
et al. (2004). NASS data were published by the USDA
National Agricultural Statistics Service (2004).
References
Lubowski, R.N., M. Vesterby, S. Bucholtz, A. Baez, and
MJ. Roberts. 2006. Major uses of land in the United
States, 2002. Economic Information Bulletin No.
(EIB-14). U.S. Department of Agriculture, Economic
Research Service.
Smith, W.B., P.O. Miles, J.S. Vissage, and S.A. Pugh.
2004. Forest resources of the United States, 2002. USDA
Forest Service.
USDA Farm Service Agency (United States Department of
Agriculture, Farm Service Agency). 2004. The Conservation
Reserve Program: Summary and enrollment statistics, 2004.
USDA NASS (United States Department of Agriculture,
National Agricultural Statistics Service). 2004. 2002 census
of agriculture, United States summary and state data.
Report AC-02-A-51. (QA/QC);
USDA NRCS (United States Department of Agriculture,
Natural Resources Conservation Service). 2007. National
Resources Inventory, 2003 annual NRI: Land use.
USDA NRCS. 2004. National resources inventory: 2002
annual NRI.
80
-------
Urbanization and Population Change
The total number of people and their distribution on the
landscape can affect the condition of the environment
in many ways. Increasing population often means increased
urbanization, including conversion of forest, farm, and other
lands for housing, transportation, and commercial purposes.
In recent years, many communities in the U.S. have seen
an increase in developed land (residential, commercial,
industrial, and transportation uses) that outpaces popula-
tion growth. This pattern is of concern for numerous health
and environmental reasons (Frumkin et al., 2004). For
example, studies indicate that when land consumption rates
exceed the rate of population growth, per capita air pollut-
ant emissions from driving tend to be higher. Urbanization
and population growth also tend to increase the amount of
impervious surfaces and the quantity and types of products
that humans produce, use, and discard, thereby affect-
ing waste generation and management, water quality, and
chemical production and use.
The information presented in this indicator is based on
population data collected and analyzed on a decadal basis
by the U.S. Census Bureau—as well as annual "intercen-
sal" population estimates—and data collected by the U.S.
Department of Agriculture Natural Resources Conserva-
tion Service's National Resources Inventory (NRI) to track
"developed" land. Between 1977 and 1997, the NRI devel-
oped estimates every 5 years on non-federal lands in the
contiguous U.S. Since 2001 the NRI has developed annual
estimates, but based on a smaller sample size. This indicator
captures trends in overall population growth for both rural
and urban populations; the amount of developed land rela-
tive to the amount of population change, nationally and by
EPA Region; and overall population density, also nationally
and by EPA Region.
What the Data Show
The U.S. population grew from a little over 4 million
people in 1790 to over 281 million in 2000; urban popu-
lation is estimated to have grown a thousandfold over
that period (Exhibit 4-8). The population nearly doubled
between 1950 and 2000.
The rates of population and developed land growth over
5-year intervals increased between 1982 and 1997, before
declining slightly between 1997 and 2002. Over all four
5-year increments, the amount of developed land increased
at nearly twice the rate of the population (Exhibit 4-9).
Between 1982 and 2003, the amount of developed land
in the U.S. in the 48 contiguous states (not including
the District of Columbia) grew by more than 35 million
acres, representing a cumulative increase of more than 48
percent. The Census Bureau estimates that during the same
period, the population of the 48 states grew by nearly 58
million people, or just over 25 percent (Exhibit 4-10).
There are substantial variations in population and devel-
opment trends in different parts of the U.S. (Exhibit 4-10).
Exhibit 4-8. Population and urbanization in the
U.S., 1790-20003
n Rural population
n Urban population
1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000
Year
Coverage: 50 states and the District of Columbia.
Data source: U.S. Census Bureau, 1993, 2004
Exhibit 4-9. Percent change in population
and developed land in the contiguous U.S.
and Hawaii, 1982-2002ab
Q- 4
2
0
1982-
1987
1987-
1992
1992-
1997
1997-
2002b
Reporting period
Change in population
Change in acreage
of developed land
Coverage: Contiguous 48 states
(excluding the District of Columbia)
and Hawaii.
bBased on changes in the NRI
inventory approach, Hawaii was not sampled in 2002. Thus, the
percent change in developed land from 1997 to 2002 is based on
the 48 contiguous states only.
Data source: U.S. Census Bureau, 1996,2002b, 2006; USDA
NRCS, 2000, 2004
Between 1982 and 2003, the growth rates for developed
land were higher than population growth rates in every
region except Region 8. The largest rate of increase in
81
-------
Urbanization and Population Change
Exhibit 4-10. Percent change in population and
developed land in the contiguous U.S. by EPA
Region, 1982-20033
90
80
70
o
g> 60
| 50
g 40
o
«£ 30
20
10
Change in population
Change in acreage
of developed land
R1 R2 R3 R4 R5 R6 R7
EPA Region
Coverage: Contiguous 48 states
(excluding the District of Columbia).
Data source: U.S. Census Bureau,
1996, 2002b, 2006; USDA NRCS,
2000, 2007
R9 R10 All
U.S.
EPA Regions
population between 1982 and 2003 occurred in Region 9,
•where population increased by more than 46 percent (nearly
14 million people). Developed land in Region 9 increased by
51 percent (more than 2.8 million acres). Region 4 had the
largest rate of increase in developed land (nearly 80 percent)
and the largest absolute increases in both population (15.4
million) and developed land (11.8 million acres).
Although growth rates of population and developed land
•were high in most Regions, population density varies signif-
icantly from one Region to the next (Exhibit 4-11). In 2005,
EPA Region 2 was the most densely populated Region, at
512 people per square mile; EPA Region 10 was the least
densely populated, with an average of approximately 15 peo-
ple per square mile (including Alaska). The national average
in 2005 \vas 83.8 people per square mile.
Indicator Limitations
Census data:
Intercensal figures are estimates based on administrative
records of births, deaths, and migration, and thus differ
from the decennial census data in methodology
and accuracy.
• Sampling and non-sampling errors exist for all Census
data as a result of errors that occur during the data col-
lection and processing phases of the census.
Exhibit 4-11. Population density in the U.S. by EPA Region, 1950-20053
a
.£•
0.
O
D-
600
500
400
300
200
100
0
600
500
400
300
200
100
0
Region 1
Region 2
Region 3
Region 4
_
„
Region 5
Region 6
'50'60 70'80'90'00'05 '50'60 70'80'90'00'05 '50'60 70'80'90'00'05 '50'60 70'80'90'00'05 '50'60 70'80'90'00'05 '50'60 70'80'90'00'05
Region 7
Region 8
Region 9
Region 10
— i — i — • — i — i — • — i —
All U.S.
•••••••
EPA Regions
'50'60 70'80'90'00'05 '50'60 70'80'90'00'05 '50'60 70'80'90'00'05 '50'60 70'80'90'00'05 '50'60 70'80'90'00'05
Year
Coverage: 50 states and the District of Columbia.
Data source: U.S. Census Bureau, 2002a,c; 2006
82
-------
INDICATO
Urbanization and Population Change
• Puerto Rico and Virgin Islands data are not available for
all years, and thus have not been included. This affects
the accuracy of the statistics for Region 2.
• The criteria for estimating urban population have
changed over time as defined by the Census Bureau.
NRI data:
• NRI sampling procedures changed in 2000 to an
annual survey of fewer sample sites than had previ-
ously been sampled (starting in 1977, the NRI sampled
800,000 points every 5 years). Fewer sample points mean
increased variance and uncertainty.
• The NRI collects some data across the entire nation,
including Puerto Rico and the Virgin Islands. Land use
statistics, however, are not reported on federal lands or
for Alaska and the District of Columbia. In Exhibit 4-10,
Hawaii is also excluded.
Data Sources
Urban and rural population data for Exhibit 4-8 were
obtained from two U.S. Census Bureau publications: data
from 1790 to 1990 are from U.S. Census Bureau (1993);
2000 data are from U.S. Census Bureau (2004).
In Exhibit 4-9, population change was calculated from
annual population estimates published in U.S. Census
Bureau (1996, 2002b, 2006) (estimates for 1982/1987,
1992/1997, and 2002, respectively). Changes in acreage of
developed land were calculated based on acreage figures
originally reported every 5 years by the NRI and now
reported annually. NRI data were obtained from two pub-
lications (USDA NRCS, 2000, 2004) (1982-1997 and 2002
data, respectively).
Exhibit 4-10 is based on annual population estimates
by state, published in U.S. Census Bureau (1996, 2002b,
2006), and NRI-developed land estimates by state, pub-
lished in USDA NRCS (2000, 2007). The figure was
developed by grouping the published state data by EPA
Region, then calculating percent change between 1982
and 2003.
Population density by EPA Region (Exhibit 4-11) was
calculated based on three published data sets: population
every 10 years from 1900 to 2000 by state (U.S. Census
Bureau, 2002a); population estimates for 2005 by state
(U.S. Census Bureau, 2006); and land area by state (U.S.
Census Bureau, 2002c).
References
Frumkin, H., L. Frank, and R. Jackson. 2004. Urban
sprawl and public health: Designing, planning, and building
for healthy communities. Washington, DC: Island Press.
U.S. Census Bureau. 2006. Annual estimates of the popula-
tion for the United States, Regions, states, and for Puerto
Rico: April 1, 2000 to July 1, 2006 (NST-EST2006-01)
released December 22, 2006. Washington, DC. Available from
U.S. Census Bureau. 2004. Statistical abstract of the United
States 2004-2005: The national data book. Washington DC.
U.S. Census Bureau. 2002a. Demographic trends in the
20th century: Census 2000 special reports. Washington,
DC.
U.S. Census Bureau. 2002b. Time series of intercensal state
population estimates: April 1, 1990 to April 1, 2000. Table
CO-EST2001-12-00. Washington, DC.
U.S. Census Bureau. 2002c. Population, housing units,
area, and density for states, 2000. Washington, DC.
U.S. Census Bureau. 1996. Intercensal estimates of
the total resident population of states: 1980 to 1990.
Washington, DC.
U.S. Census Bureau. 1993. 1990 census of population
and housing: population and unit counts, United States.
1990-CPH-2-1. Washington, DC.
USDA NRCS (United States Department of Agriculture,
Natural Resources Conservation Service). 2007. National
Resources Inventory, 2003 annual NRI: Land use.
USDA NRCS. 2004. National resources inventory: 2002
annual NRI.
USDA NRCS. 2000. Summary report: 1997 national
resources inventory (revised December 2000). Washington,
DC and Ames, IA: USDA Natural Resource Conservation
Service,
83
-------
INDICATO
Fertilizer Applied for Agricultural Purposes
Commercial fertilizers are applied to agricultural crops
to increase crop yields. Prior to the 1950s, most farm-
ing occurred on small family farms with limited use of
chemicals. The shift since then to larger corporate farms
has coincided with the use of chemical fertilizers in mod-
ern agricultural practices. The three major types of com-
mercial fertilizer used in the U.S. are nitrogen, phosphate,
and potash.
Nitrogen (N) is found primarily in the organic form
in soils, but can also occur as nitrate. Because nitrate is
extremely soluble and mobile, it can lead to nuisance algal
growth, mostly in downstream estuaries, and cause con-
tamination of drinking water. Phosphorus (P) occurs in soil
in several forms, both organic and inorganic. Phosphorus
loss due to erosion is common and phosphate, while less
soluble than nitrate, can easily be transported in runoff.
Phosphorus/phosphate runoff can lead to nuisance algae
and plant growth, often in freshwater streams, lakes, and
estuaries. Potash is the oxide form of potassium (K) and its
principal forms as fertilizer are potassium chloride, potas-
sium sulfate, and potassium nitrate. When used at recom-
mended application rates, there are few to no adverse effects
from potassium, but it is a common component of mixed
fertilizers used for high crop yields and is tracked in the
fertilizer use surveys conducted.
This indicator shows use of the three major fertilizers in
pounds per acre of land per year (expressed as N, P, or K)
used for crop production from 1960 to 2005. Data are from
an annual survey for agricultural crops conducted by the
U.S. Department of Agriculture (USDA) National Agri-
cultural Statistics Service (NASS) and from the Economic
Research Service (ERS) Major Land Use series. Acre-
age used for crop production includes cropland harvested
and crop failure as estimated in the ERS series. Cropland
estimates as used in this indicator are a subset of agricul-
tural land estimates discussed in the Land Cover and Land
Use indicators. NASS also produces an annual Agricultural
Chemical Usage report on four to five targeted field crops,
based on data compiled from the Agricultural Resources
Management Survey (ARMS). The ARMS surveys farmers
in major agriculture-producing states that together account
for a large percentage of crop acreage for corn, soybeans,
cotton, and wheat. Results are presented for the years
2005-2006 by EPA Region.
What the Data Show
Based on fertilizer sales data, total use of the three major
commercial fertilizers has steadily increased, from 46.2
nutrient pounds per acre per year (Ibs/acre/yr) in 1960 to
138 Ibs/acre/yr in 2005, an increase of 199 percent (Exhibit
4-16). During this period, cropland used for crop produc-
tion generally has fluctuated between 290 and 360 million
acres with the largest changes occurring between 1969
(292 million acres) and 1981 (357 million acres) (Lubowski
Exhibit 4-16. Commercial fertilizer use in the
U.S., 1960-2005a
— 160
I 140
o
3 120
o
! | 100
! £ 80
: Q-
: "i GO
o
£• 4°
c
-------
Fertilizer Applied for Agricultural Purposes
Production of winter, durum, and other spring wheat
occurred on about 57 million acres in 2006 and is distrib-
uted across EPA Regions 5, 6, 7, 8, and 10. Wheat typically
accounts for about 10 percent of all commercial fertilizer
used (Daberkow and Huang, 2006).
Soybeans were the fastest-growing crop in total acreage,
increasing from 57.8 million acres in 1990 to 75.5 mil-
lion acres in 2006 (USDA NASS, 2007c). The majority of
soybean acreage (80 percent) is concentrated in the upper
Midwest in EPA Regions 5 and 7. Soybeans require the
least fertilizer per acre of the four crops described here.
Overall, production of these four crops in the ARMS
states used slightly more than 13.25 million tons per year
(MT/yr) of fertilizer in 2005-2006 (Exhibit 4-17) of the
21.7 MT/yr estimated (2005-2006 average) by ERS for all
crops produced in the entire U.S. Of this amount, slightly
less than half (5.8 MT/yr) was applied in EPA Region 5
(Exhibit 4-17), most of which was used for corn. An addi-
tional 3.7 MT/yr was applied in EPA Region 7, primarily
on corn or soybeans.
Indicator Limitations
• USDA national estimates of fertilizer use are based on
sales data provided by states, not actual fertilizer usage,
and are susceptible to differing reporting procedures or
accuracy from state to state.
• Data to identify cropland used for crop production are
from the major land use series discussed in the Land
Cover and Land Use indicators and do not include
Alaska and Hawaii.
• Within the ARMS, not all states report fertilizer data
every year for each crop type, making it difficult to
establish year-to-year trends (a decrease in fertilizer use
for a specific crop might be attributed to failure of a state
to report, rather than an actual decrease of use).
• ARMS sampling is limited to program states, which
represent 82 to 99 percent of crop acreage (across all
surveyed crops) for the years 2005 and 2006, depending
on crop type.
• The NASS Acreage report has estimates of acreage in
production for the entire nation by crop, while fertilizer
sales data are based only on USDA program states. Even
though USDA program states represent the majority of
U.S. planted acreage (often over 90 percent), the abil-
ity to generalize the data to the country as a whole is
unknown, as non-program states, while representing a
small percentage of a crop, might have much different
application rates due to climate, weather, etc.
Fertilizer applied to trees that are considered agricultural
crops (e.g., nut-producing trees) is included in field crop
summaries, but fertilizer applied in silviculture (e.g.,
Exhibit 4-17. Fertilizer use for four common
crops (corn, cotton, soybeans, and wheat) in
major agriculture-producing states, by EPA
Region, 2005-20063
n D h
n Phosphate
n Nitrogen
E
^
i — i
R2 R3 R4 R5 R6 R7
EPA Region
R9 R10
"Coverage: States surveyed by EPA Regions
USDA's Agricultural Resource
Management Survey (ARMS)
Program in 2005-2006 for corn,
cotton, soybeans, and wheat. Each
commodity was surveyed in a
different subset of states, which
together account for a substantial
portion of the nation's production of
that particular commodity. No states in Region 1 were surveyed
by the ARMS Program for corn, cotton, soybeans, or wheat.
Data source: USDA NASS, 2006b, 2007b
southern pine plantations) is not covered by the NASS
data collection system.
Loading of nutrients in aquatic systems is not necessarily
correlated directly with fertilizer use, but rather with the
levels of fertilizer applied in excess of amounts used by
crops, natural vegetation, and soil biota.
Data Sources
Exhibit 4-16 is based on two sets of summary data
from ERS. Annual estimates of fertilizer use from 1960
through 2005, by nutrient, were obtained from Wiebe
and Gollehon (2006) (see summary tables, http://www.
ers.usda.gov/Data/FertilizerUse/). Fertilizer use per acre
\vas calculated based on annual estimates of the amount
of cultivated (harvested or failed) cropland from 1960 to
2005 published in Lubowski et al. (2006) (see summary
tables, http://www.ers.usda.gov/Data/MajorLandUses/
MLUsummarytables.pdf).
Exhibit 4-17 is based on fertilizer use data from USDA's
2005 and 2006 ARMS survey, which were obtained from
USDA NASS (2006b, 2007b). The published data are by
state, so additional aggregation was required to report by
EPA Region (USDA NASS, 2001, 2004, 2005a,b, 2006a).
85
-------
INDICATO
Fertilizer Applied for Agricultural Purposes
References
Daberkow, S., and W. Huang. 2006. Nutrient manage-
ment. In: Wiebe, K., and N. Gollehon, eds. 2006. Agricul-
tural resources and environmental indicators, 2006 edition.
EIB-16. U.S. Department of Agriculture, Economic
Research Service,
Lubowski, R.N., M. Vesterby, S. Bucholtz, A. Baez, and
MJ. Roberts. 2006. Major uses of land in the United States,
2002. EIB-14. U.S. Department of Agriculture, Economic
Research Service,
USDA NASS (United States Department of Agriculture,
National Agricultural Statistics Service). 2007a. Acreage.
USDA NASS. 2007b. Agricultural chemical usage, 2006
field crop summary. May.
USDA NASS. 2007c. Crop Production Historical Track
Records,
USDA NASS. 2006a. Acreage.
USDA NASS. 2006b. Agricultural chemical usage,
2005 field crop summary. May.
USDA NASS. 2005a. Acreage.
USDA NASS. 2005b. Crop production: 2004 summary.
Cr Pr 2-1 (05).
USDA NASS. 2004. Acreage.
USDA NASS. 2001. Agricultural chemical usage, 2000
field crops summary,
Wiebe, K., and N. Gollehon, eds. 2006. Agricultural
resources and environmental indicators, 2006 edition.
EIB-16. U.S. Department of Agriculture, Economic
Research Service,
86
-------
INDICATOI
Cardiovascular Disease Prevalence and Mortality
The broad category of cardiovascular disease (CVD)
includes any disease involving the heart and blood
vessels. Coronary heart disease, cerebrovascular disease
(commonly known as stroke), and hypertension are the
major cardiovascular diseases (American Heart Association,
2007). In addition to being a major risk factor for heart
disease and stroke, hypertension is a commonly diagnosed
disease that can also lead to kidney damage and other
health problems. Obesity, physical inactivity, and sodium
intake are all important risk factors for hypertension (NIH,
2004). Since 1900, CVD has been the leading cause of
death in the U.S. every year except 1918 (American Heart
Association, 2007) (General Mortality indicator). The
U.S. age-adjusted mortality rate for CVD reached a peak
in 1950 (CDC, 1999). Between 1950 and 1999, the age-
adjusted mortality rate for CVD declined 60 percent. The
major risk factors for CVD include tobacco use, high blood
pressure, high blood cholesterol, diabetes, physical inactiv-
ity, and poor nutrition (CDC, 2004; American Heart Asso-
ciation, 2007).
Environmental exposures may also play a role in CVD
morbidity and mortality independent of other risk factors.
However, susceptible populations such as the elderly and
other high-risk populations may be most impacted. For
example, studies have shown exposure to ambient air-
borne particulate matter to be associated with increased
hospitalizations and mortality among older individu-
als, largely due to cardiopulmonary and cardiovascular
disease (U.S. EPA, 2004). Environmental tobacco smoke
(ETS) may also contribute to CVD. Although the smoke
to which a nonsmoker is exposed is less concentrated
than that inhaled by smokers, research has demonstrated
increased cardiovascular-related health risks associated
with ETS (State of California, 2005).
This indicator presents U.S. adult (age 18 and older)
prevalence rates for heart disease (all types), coronary heart
disease, stroke, and hypertension; and mortality rates for
CVD as a whole as well as coronary heart disease (includ-
ing myocardial infarction), stroke, and hypertension. CVD
prevalence data were compiled between 1997 and 2006
from the National Health Interview Survey (NHIS), con-
ducted by the Centers for Disease Control and Prevention's
(CDC's) National Center for Health Statistics (NCHS).
The NHIS is the principal source of information on the
health of the civilian non-institutionalized population of
the U.S. and since 1960 has been one of the major data col-
lection programs of NCHS. CVD prevalence is based on
the number of adults who reported that they had ever been
told by a doctor or other health practitioner that they had
a specified CVD. Mortality data (all ages) were compiled
between 1979 and 2004 using the National Vital Statis-
tics System (NVSS), maintained by NCHS. The NVSS
Exhibit 5-23. Cardiovascular disease
prevalence in U.S. adults (age 18 and older),
1997-20063
250
200
S
°-150
o
— 100
CD
ro
CC
50
'97 '98 '99 '00 '01 '02 '03 '04 '05 '06
Year
aRates presented are crude rates.
Data source: NCHS, 1999-2005, 2006a,b,
2007
Exhibit 5-24. Age-adjusted cardiovascular
disease mortality rates in the U.S., 1979-2004at
IB
'85
'90
'95
'00
Year
aDue to differences in the ICD system used for
classifying mortality, data from 1979-1998
should not be directly compared to data from
1999-2004 [ICD-9 codes: 390-434, 436-448
(1979-1998); ICD-10 codes: IOO-I78
(1999-2004)].
"Rates are age-adjusted to the 2000 U.S.
standard population.
Data source: CDC, 2007
registers virtually all deaths and births nation-wide, with
data coverage from 1933 to 2004 and from all 50 states and
the District of Columbia.
87
-------
INDICATOI
Cardiovascular Disease Prevalence and Mortality
What the Data Show
CVD Prevalence
Among adults 18 years and older, the prevalence of heart
disease and stroke between 1997 and 2006 has remained
essentially the same (Exhibit 5-23). In contrast, the preva-
lence of hypertension has shown an increase from 191.6
cases per 1,000 in 1999 to 234.1 cases per 1,000 in 2006.
Gender, race, and age differences in CVD prevalence
exist. The prevalence of coronary heart disease is consis-
tently higher among males than among females (74.1 cases
per 1,000 compared with 54.2 cases per 1,000 for women
in 2006). In contrast, hypertension is more prevalent
among women (238.4 cases per 1,000 for women compared
•with 229.5 for men in 2006). Among the racial groups
reported, American Indians and Alaska Natives typi-
cally had the highest prevalence of coronary heart disease
between 1999 and 2003. In 2006, however, whites had the
highest prevalence of coronary heart disease (67.8 cases per
1,000), followed by American Indians and Alaska Natives
(55.5 cases per 1,000), blacks or African Americans (52.0
cases per 1,000), and Asians (28.6 cases per 1,000). In 2006,
Asians also consistently had the lowest prevalence of stroke
(13.8 cases per 1,000) and hypertension (157.0 cases per
1,000) among the racial groups reported. In addition, the
Hispanic or Latino population had a consistently lower
prevalence of the major CVD-related diseases com-
pared with the non-Hispanic or Latino population from
1999-2006, the period for which these data are available.
For example, in 2006, prevalence in Hispanics or Latinos
\vas lower than in non-Hispanics or Latinos for coronary
heart disease (31.7 versus 68.6 cases per 1,000, respec-
tively), hypertension (147.5 versus 247.0 cases per 1,000,
respectively), and stroke (12.2 versus 27.6 cases per 1,000,
respectively). (Data not shown.)
CVD Mortality
In 1998, the national age-adjusted CVD mortality rate (all
types) was 352.0 per 100,000 compared to a rate of 541.0
per 100,000 in 1980 (Exhibit 5-24). This decline appears
to continue after 1999, with the rate dropping from 349.3
per 100,000 in 1999 to 286.5 per 100,000 in 2004. Both
coronary heart disease and stroke mortality rates have been
declining in the U.S. The age-adjusted coronary heart
disease mortality rate ranged from 345.2 per 100,000 in
1980 to 197.1 per 100,000 in 1998. For stroke mortality, the
age-adjusted rate ranged from 97.1 per 100,000 in 1979 to
59.3 per 100,000 in 1998. The age-adjusted mortality rates
for myocardial infarction ranged from 157.9 in 1979 to 76
per 100,000 in 1998. The age-adjusted mortality rates for
coronary heart disease, stroke, and myocardial infarction in
2004 were 150.2, 50.0, and 52.3 per 100,000, respectively,
compared to 194.6, 61.6, and 73.2 per 100,000, respectively,
Exhibit 5-25. Age-adjusted coronary heart
disease mortality rates in the U.S. by EPA
Region, 1979-2004ab
450
400
350
.3^300
S ° 250
ti
1-12°°
< 150
100
50
0
-R1
R2
R3
R4
-R5
R6
R7
R8
-R9
-R10
-Nat'l
'90 '95
Year
'00
aDue to differences in the ICD
system used for classifying
mortality, data from 1979-1998
should not be directly compared
to data from 1999-2004 [ICD-9
codes: 410-414, 429.2
(1979-1998); ICD-10 codes:
I20-I25 (I999-2004)].
bRates are age-adjusted to the
2000 U.S. standard population.
Data source: CDC, 2007
in 1999. Death rates from hypertension remained essentially
the same between 1999 and 2004.
Both coronary heart disease and stroke mortality have
been declining over time in each of the 10 EPA Regions
(Exhibits 5-25 and 5-26). In 1979, coronary heart disease
and stroke age-adjusted mortality rates ranged from 285.6
(Region 10) to 401.9 (Region 2) per 100,000 and 80.3
(Region 2) to 111.4 (Region 4) per 100,000, respectively.
In 1998, coronary heart disease and stroke mortality rates
ranged from 145.6 (Region 8) to 233.2 (Region 2) per
100,000 and 43.2 (Region 2) to 68.5 per (Region 10)
100,000, respectively. The observed decreases in coronary
heart disease and stroke mortality also appear to continue
in the 1999-2004 period.
Differences exist in CVD mortality rates among gender,
racial, and age groups. For example, in 2004, those age 65
and older had the highest CVD (all types), coronary heart
disease, and stroke mortality (1,898.7, 990.8, and 346.2 per
100,000, respectively). For the same year, the age-adjusted
CVD, coronary heart disease, and stroke mortality rates for
those 45 to 64 years of age were 172.7, 98.5, and 22.5 per
-------
INDICATOI
Cardiovascular Disease Prevalence and Mortality
100,000, respectively. Notable differences in CVD (all types)
and, specifically, coronary heart disease mortality rates
exist between males and females, but not for stroke mortal-
ity. Coronary heart disease mortality among males in 2004
was 194.2 per 100,000, compared to 116.7 per 100,000 for
•women. In 2004, black or African American males had the
highest CVD mortality rate at 451.1 per 100,000 compared
to white males (333.6 per 100,000), black or African Ameri-
can females (331.0 per 100,000), and white females (236.7
per 100,000). (Data not shown.)
Indicator Limitations
• Prevalence data reported in the NHIS are based on
self-reported responses to specific questions pertaining
to CVD-related illnesses, and are subject to the biases
associated with self-reported data. Self-reported data can
underestimate the disease prevalence being measured if,
for -whatever reason, the respondent is not fully aware of
his/her condition.
• All prevalence data are based on crude rates and are not
age-adjusted, as CDC did not report age-adjusted data
prior to 2002 in the data sources used for this indicator.
Therefore, the reported disease prevalence rates across
time or within different race and gender subgroups
may not reflect differences in the age distribution of the
populations being compared.
• For one or more years for which data are presented,
coronary heart disease and stroke prevalence rates pre-
sented for Native Americans and Alaska Natives have
a relative standard error of greater than 30 percent. In
addition, stroke prevalence rates for one or more years
for \vhich data are presented for Asians have a relative
standard error of greater than 30 percent. As such, these
rates should be used -with caution as they do not meet the
standard of reliability or precision.
• CVD mortality rates are based on underlying cause of
death as entered on a death certificate by a physician.
Some individuals may have had competing causes of
death. "When more than one cause or condition is entered
by the physician, the underlying cause is determined by
the sequence of conditions on the certificate, provisions
of the ICD [International Classification of Diseases], and
associated selection rules and modifications" (CDC, n.d.).
Consequently, some misclassification of reported mortal-
ity might occur in individuals -with competing causes of
death, as -well as the possible underreporting of CVD as
the cause of death.
• The International Classification of Diseases 9th Revision
(ICD-9) codes -were used to specify underlying cause of
death for years 1979-1998. Beginning in 1999, cause of
death is specified -with the International Classification of
Diseases 10th Revision (ICD-10) codes. The two revisions
Exhibit 5-26. Age-adjusted stroke mortality
rates in the U.S. by EPA Region, 1979-20043
120
100
1°
60
40
20
—R1
—R2
—R3
R4
-R5
—R6
R7
R8
—R9
-R10
—Nat'l
'85
'90
'95
'00
Year
aDue to differences in the ICD
system used for classifying
mortality, data from 1979-1998
should not be directly compared
to data from 1999-2004 [ICD-9
codes: 430-434, 436-438
(1979-1998); ICD-10 codes:
I60-I69 (1999-2004)].
bRates are age-adjusted to the
2000 U.S. standard population.
Data source: CDC, 2007
EPA Regions
differ substantially, and to prevent confusion about the
significance of any specific disease code, data queries are
separate.
Data Sources
CVD prevalence data -were obtained from annual reports
published by NCHS (NCHS, 1999-2007), which summa-
rize health statistics compiled from the NHIS (http://www
cdc.gov/nchs/products/pubs/pubd/series/ser.htm). CVD
mortality statistics -were obtained from CDC's "compressed
mortality" database, accessed through CDC WONDER
(CDC, 2007) (http://wonder.cdc.gov/mortSQL.html). EPA
Regional mortality statistics -were generated by combining
and age-adjusting state-by-state totals for each EPA Region
using data from CDC WONDER.
References
American Heart Association. 2007. Heart disease and
stroke statistics—2007 update. A report from the American
Heart Association Statistics Committee and Stroke Statis-
tics Subcommittee. Circulation (115):e69-el71.
89
-------
INDICATOR
Cardiovascular Disease Prevalence and Mortality
CDC (Centers for Disease Control and Prevention). 2007.
CDC Wide-ranging OnLine Data for Epidemiologic
Research (WONDER). Compressed mortality file, under-
lying cause of death. 1999-2004 (with ICD 10 codes) and
1979_1998 (with ICD 9 codes). Accessed September 2007.
CDC. 2004. The burden of chronic diseases and their risk
factors—national and state perspectives,
CDC. 1999. Decline in deaths from heart disease and
stroke, United States, 1990-1999. Washington, DC.
CDC. n.d. CDC WONDER: Help page for compressed
mortality file. Accessed September 2007.
NCHS (National Center for Health Statistics). 2007.
Summary health statistics for U.S. adults: National Health
Interview Survey, 2006. Vital Health Stat. 10(235).
NCHS. 2006a. Summary health statistics for U.S. adults:
National Health Interview Survey, 2005. Vital Health Stat.
10(232).
NCHS. 2006b. Summary health statistics for U.S. adults:
National Health Interview Survey, 2004. Vital Health Stat.
10(228).
NCHS. 2005. Summary health statistics for U.S. adults:
National Health Interview Survey, 2003. Vital Health Stat.
10(225).
NCHS. 2004. Summary health statistics for U.S. adults:
National Health Interview Survey, 2002. Vital Health Stat.
10(222).
NCHS. 2003. Summary health statistics for U.S. adults:
National Health Interview Survey, 2001. Vital Health Stat.
10(218).
NCHS. 2002. Summary health statistics for U.S. adults:
National Health Interview Survey, 2000. Vital Health Stat.
10(215).
NCHS. 2001. Summary health statistics for U.S. adults:
National Health Interview Survey, 1999. Vital Health Stat.
10(212).
NCHS. 2000. Summary health statistics for U.S. adults:
National Health Interview Survey, 1998. Vital Health Stat.
10(209).
NCHS. 1999. Summary health statistics for U.S. adults:
National Health Interview Survey, 1997. Vital Health Stat.
10(205).
NIH (National Institute of Health). 2004. NIH news: The
increasing number of adults with high blood pressure.
State of California. 2005. Proposed identification of envi-
ronmental tobacco smoke as a toxic air contaminant. Part B:
Health effects assessment for environmental tobacco smoke.
As approved by the Scientific Review Panel on June 24,
2005. California Environmental Protection Agency, Office
of Environmental Health Hazard Assessment.
U.S. EPA (United States Environmental Protection Agency).
2004. Air quality criteria for particulate matter. Volumes
I (EPA/600/P-99/002aF) and II (EPA/60O/P-99/002bF).
National Center for Environmental Assessment—RTF
Office, Office of Research and Development.
90
-------
INDICATOI
Chronic Obstructive Pulmonary Disease Prevalence and Mortality
Chronic obstructive pulmonary disease (COPD), some-
times referred to as chronic lung disease, is a disease
that damages lung tissue or restricts airflow through the
bronchioles and bronchi (NHLBI, 2003). Chronic bron-
chitis and emphysema are the most frequently occurring
COPDs. Smoking is the most common cause of COPD,
including cigarette, pipe, and cigar smoking (NHLBI,
2003). Other risk factors in the development and progres-
sion of COPD include asthma, exposure to air pollutants
in the ambient air and -workplace environment, genetic
factors, and respiratory infections (CDC, 2003; American
Lung Association, 2004).
Environmental tobacco smoke (ETS) may also increase
the risk of developing COPD. The effect of chronic ETS
exposure alone on pulmonary function in otherwise healthy
adults is likely to be small. However, in combination with
other exposures (e.g., prior smoking history, exposure to
occupational irritants or ambient air pollutants), ETS expo-
sure could contribute to chronic respiratory impairment.
Children are especially sensitive to the respiratory effects of
ETS exposure (State of California, 2005).
This indicator presents U.S. adult (age 18 and older)
prevalence rates for chronic bronchitis and emphysema
and mortality rates for COPD as a whole and for chronic
bronchitis and emphysema. COPD prevalence data were
compiled from 1999 to 2006 from the National Health
Interview Survey (NHIS), conducted by the Centers for
Disease Control and Prevention's (CDC's) National Center
for Health Statistics (NCHS). The NHIS is the principal
source of information on the health of the civilian non-
institutionalized population of the U.S. and since 1960 has
been one of the major data collection programs of NCHS.
COPD prevalence is based on the number of adults who
reported that they had ever been told by a doctor or
other health practitioner that they had chronic bronchitis
or emphysema. Mortality data (all ages) were compiled
between 1979 and 2004 using the National Vital Statistics
System (NVSS), maintained by NCHS. The NVSS reg-
isters virtually all deaths and births nation-wide, -with data
coverage from 1933 to 2004 and from all 50 states and the
District of Columbia.
What the Data Show
COPD Prevalence
Exhibit 5-27 presents the prevalence of chronic bronchitis
(panel A) and emphysema (panel B) from 1999 to 2006.
The reported total prevalence of chronic bronchitis in
U.S. adults over the age of 18 years ranged from a low of
40 (2003) to a high of 55 (2001) cases per 1,000. A small
increase in prevalence of chronic bronchitis can be seen
from 1999 to 2001, -with a subsequent overall decline from
2001 to 2006. The reported total prevalence of emphysema
in U.S. adults during the same time period ranged from 14
(1999) to 18 (2006) cases per 1,000. No notable change in
the prevalence for emphysema -was evident during this time
period. Exhibit 5-27 also displays chronic bronchitis and
emphysema prevalence by race. Chronic bronchitis preva-
lence -was higher among -white (designated as "-white only")
adults than black ("black or African American only")
adults during 1999 (46 versus 36 cases per 1,000, respec-
tively), 2000 (49 versus 40 cases per 1,000, respectively),
and 2004 (44 versus 36 cases per 1,000, respectively). How-
ever, in 2006 rates in black and -white adults are the same
(43 cases per 1,000). Throughout the entire time period,
emphysema prevalence is consistently higher among -white
adults than black adults.
In addition, the Hispanic or Latino population had a
consistently lower prevalence of chronic bronchitis and
emphysema diseases than the non-Hispanic or Latino
population from 1999-2006, the period for -which these
Exhibit 5-27. Chronic bronchitis and emphysema prevalence in U.S. adults (age 18 and older) by
race, 1999-2006a
A. Chronic bronchitis
B. Emphysema
aRates presented are
crude rates.
Data source: NCHS,
2001-2005, 2006a,b,
2007
Year
91
-------
INDICATOI
Chronic Obstructive Pulmonary Disease Prevalence and Mortality
data are available. For example, in 2006, prevalence in His-
panics or Latinos was lower than non-Hispanics or Latinos
for chronic bronchitis (22 compared to 46 cases per 1,000,
respectively) and emphysema (4 compared to 21 cases per
1,000, respectively). (Data not shown.)
Gender differences are also seen. In 2006, females had
about twice the reported prevalence of chronic bronchitis
than males (57 versus 27 cases per 1,000 respectively), a
consistently observed difference between 1997 and 2006.
Unlike with chronic bronchitis, the prevalence rates for
emphysema have been consistently higher in males than in
females. (Data not shown.)
COPD Mortality
In 2004, COPD continues to be the fourth leading cause
of mortality, accounting for 121,987 (5.1 percent) of all
deaths (General Mortality indicator). The age-adjusted
mortality rate for COPD as a whole has increased over
time, with rates ranging from 25.5 per 100,000 in 1979 to
41.8 per 100,000 in 1998. From 1999 to 2004, rates held
steadier, ranging from 45.4 per 100,000 in 1999 to 41.1 per
100,000 in 2004. Mortality rates for emphysema (6.9 and
6.5 per 100,000 for 1979 and 1998, respectively, and 6.5
and 4.6 per 100,000 for 1999 and 2004, respectively) and
chronic bronchitis (1.7 and 0.9 per 100,000 for 1979 and
1998, respectively, and 0.2 and 0.1 per 100,000 for 1999
and 2004, respectively) have not changed substantially dur-
ing the same time period. (Data not shown.)
Exhibit 5-28 presents the overall COPD mortality
rates in the U.S. and the 10 EPA Regions for 1979-1998
and 1999-2004. The age-adjusted COPD mortality rates
have been increasing in each of the 10 Regions from
1979 to 1998. The rates ranged from 22.2 (Region 2) to
31.2 (Region 8) per 100,000 in 1979 and 33.5 (Region
2) to 47.9 (Region 8) per 100,000 in 1998. Between 1999
and 2004, COPD mortality rates in each of the 10 EPA
Regions have generally declined.
COPD age-adjusted mortality rates have been declin-
ing for males over time, with a rate of 58.7 per 100,000 in
1999 compared to 49.5 per 100,000 in 2004. For females,
the rates are lower than males and have been relatively
stable between 1999 and 2004 (37.7 and 36.0 per 100,000,
respectively). The COPD age-adjusted mortality rate is
higher among whites (43.2 per 100,000 in 2004) com-
pared to blacks or African Americans (28.2 per 100,000 in
2004). COPD mortality rate increases with age: the 2004
rates were 0.3, 1.1, 21.0, and 284.3 per 100,000 for those
age 0-14 years, 15-44 years, 45-64 years, and 65 years and
older, respectively. (Data not shown.)
Indicator Limitations
• Prevalence data presented in the NHIS are based on
self-reported responses to specific questions pertaining
to COPD-related illnesses, and are subject to the biases
Exhibit 5-28. Age-adjusted chronic obstructive
pulmonary disease mortality rates in the U.S. by
EPA Region, 1979-2004ab
'00
aDue to differences in the ICD
system used for classifying
mortality, data from 1979-1998
should not be directly compared
to data from 1999-2004 [ICD-9
codes: 490-494, 496
(1979-1998); ICD-10 codes:
J40-J47 (1999-2004)].
bRates are age-adjusted to the
2000 U.S. standard population.
Data source: CDC, 2007
EPA Regions
associated with self-reported data. Self-reported data can
underestimate the disease prevalence being measured if,
for -whatever reason, the respondent is not fully aware of
his/her condition.
All prevalence data are based on crude rates and are not
age-adjusted, as CDC did not report age-adjusted data
prior to 2002 in the data sources used for this indicator.
Therefore, the reported disease prevalence rates across
time or within different race and gender subgroups
may not reflect differences in the age distribution of the
populations being compared.
COPD mortality rates are based on underlying cause
of death as entered on a death certificate by a physi-
cian. Some individuals may have had competing causes
of death. "When more than one cause or condition is
entered by the physician, the underlying cause is deter-
mined by the sequence of conditions on the certificate,
provisions of the ICD [International Classification of
Diseases], and associated selection rules and modifica-
tions" (CDC, n.d.). Consequently, some misclassifica-
tion of reported mortality might occur in individuals
92
-------
INDICATOI
Chronic Obstructive Pulmonary Disease Prevalence and Mortality
•with competing causes of death, as well as the possible
underreporting of COPD as the cause of death.
• The International Classification of Diseases 9th Revision
(ICD-9) codes were used to specify underlying cause of
death for years 1979-1998. Beginning in 1999, cause of
death is specified with the International Classification of
Diseases 10th Revision (ICD-10) codes. The two revi-
sions differ substantially, and to prevent confusion about
the significance of any specific disease code, data queries
are separate.
Data Sources
COPD prevalence data were obtained from annual reports
published by NCHS (NCHS, 2001-2005, 2006a,b, 2007),
•which summarize health statistics compiled from the NHIS
(http://-www.cdc.gov/nchs/products/pubs/pubd/series/ser.
htm). Mortality statistics were obtained from CDC's "com-
pressed mortality" database, accessed through CDC WON-
DER (CDC, 2007) (http://wonder.cdc.gov/mortSQL.
html). EPA Regional mortality statistics were generated by
combining and age-adjusting state-by-state totals for each
EPA Region using data from CDC WONDER.
References
American Lung Association. 2004. Chronic obstruc-
tive pulmonary disease (COPD) fact sheet. Accessed
February 7, 2005.
CDC (Centers for Disease Control and Prevention). 2007.
CDC Wide-ranging OnLine Data for Epidemiologic
Research (WONDER). Compressed mortality file, under-
lying cause of death. 1999-2004 (with ICD 10 codes) and
1979_1998 (with ICD 9 codes). Accessed October 2007.
CDC. 2003. Facts about chronic obstructive pulmonary
disease (COPD). Accessed February 7, 2005.
CDC. n.d. CDC WONDER: Help page for compressed
mortality file. Accessed September 2007.
NCHS (National Center for Health Statistics). 2007.
Summary health statistics for U.S. adults: National Health
Interview Survey, 2006. Vital Health Stat. 10(235).
NCHS. 2006a. Summary health statistics for U.S. adults:
National Health Interview Survey, 2005. Vital Health Stat.
10(232).
NCHS. 2006b. Summary health statistics for U.S. adults:
National Health Interview Survey, 2004. Vital Health Stat.
10(228).
NCHS. 2005. Summary health statistics for U.S. adults:
National Health Interview Survey, 2003. Vital Health Stat.
10(225).
NCHS. 2004. Summary health statistics for U.S. adults:
National Health Interview Survey, 2002. Vital Health Stat.
10(222).
NCHS. 2003. Summary health statistics for U.S. adults:
National Health Interview Survey, 2001. Vital Health Stat.
10(218).
NCHS. 2002. Summary health statistics for U.S. adults:
National Health Interview Survey, 2000. Vital Health Stat.
10(215).
NCHS. 2001. Summary health statistics for U.S. adults:
National Health Interview Survey, 1999. Vital Health Stat.
10(212).
NHLBI (National Heart, Lung, and Blood Institute). 2003.
Chronic obstructive pulmonary disease fact sheet. NIH
publication No. 03-5229. Bethesda, MD: U.S. Department
of Health and Human Services,
State of California. 2005. Proposed identification of envi-
ronmental tobacco smoke as a toxic air contaminant. Part B:
Health effects assessment for environmental tobacco smoke.
As approved by the Scientific Review Panel on June 24,
2005. California Environmental Protection Agency, Office
of Environmental Health Hazard Assessment.
93
-------
INDICATO
Forest Extent and Type
The forests of the U.S. cover extensive lands in both the
eastern and western thirds of the country. While the
amount of forest land has remained nearly unchanged since
the beginning of the 20th century, regional changes both in
amount and types of forest cover have occurred as a result
of changing patterns of agriculture and development. The
distribution of various forest cover types is a critical deter-
minant of the condition of forest ecosystems.
This indicator is based on data from the U.S. Department
of Agriculture (USDA) Forest Service Forest Inventory and
Analysis (FIA) program. The FIA program, using a statisti-
cal survey design and comparable methods across the U.S.,
collects various data that help assess the extent, type, age,
and health of the nation's forest land. Because the surveys are
repeated over time, the FIA data provide an indication of
trends in both the extent and composition of forest land. The
extent data are collected for all forest lands across the nation,
but species composition data over time are only available for
timberland as defined by FIA data collection procedures (that
is, forests capable of producing at least 20 cubic feet per acre
per year of industrial wood and not withdrawn from timber
utilization by statute or regulation). Timberland makes up
94 percent of the forest land area in the eastern U.S. and 39
percent of forest land in the western U.S. as of 2002 (Smith et
al., 2004). Extent data are collected for individual states, but
have been summarized by EPA Region for this indicator.
What the Data Show
After a slight increase in forest land nation-wide between
1907 and 1938, forest acreage decreased by more than 16
million acres between 1938 and 1977, before increasing
by 5.3 million acres over the past three decades (Exhibit
6-2). There are variations in trends in forest cover among
the different EPA Regions. For example, between 1907
and 2002, forest land declined by roughly 22 million acres
in Region 6 and more than 12 million acres in Region 9.
Over the same period, forest land increased by 13 million
acres in Region 3 and by 10 million acres in Region 5.
In addition to changes in the extent of forest, there have
been changes in the types of forests over time (Exhibits 6-3
and 6-4). The largest changes in the eastern U.S. over the
1953-2002 period occurred in the maple-beech-birch forest
type and the oak-hickory forest type, which gained 27.5
million acres and 23 million acres, respectively, since 1953.
In the West, the fir-spruce type and Western hard-wood
type also have increased (about 11.5 million acres each) since
1953, -while the hemlock-Sitka spruce, pinyon-juniper, and
ponderosa-Jeffrey pine forest types have decreased by about
13.6 million, 8.8 million, and 8.7 million acres respectively.
The Western white pine forest type has decreased by 5.3
million acres, or about 96 percent of its 1953 acreage.
Indicator Limitations
• Data on extent of forest land have an uncertainty of 3 to
10 percent per million acres for data reported since 1953.
In 1998 Congress mandated that the FIA move to annual
inventories. While data now are collected more often,
fewer data are collected in any given year. Because area
estimates now are based on a smaller sample size, the pre-
cision of the national estimates may be reduced relative
to pre-1998 dates.
• Most of the specific data related to species and age classes
are only collected on lands classified as timberland and
not forest land in general.
• In addition to extent and species class, age class also
influences the use of forest land as habitat by different
species. Younger and older stands of forest have increased
over the past half-decade, while middle-aged stands of
more merchantable timber have decreased (Smith et al.,
2001, 2004).
Data Sources
This indicator is based on data from two USDA Forest
Service reports (Smith et al., 2001, 2004), which provide
current and historical data on forest extent and type by
state. Most data were obtained from the 2004 report; the
2001 report was consulted only for 1963 data, which were
excluded from the more recent report. Data were originally
collected by the USDA Forest Service's FIA program; origi-
nal survey data are available from the FIA database (USDA
Forest Service, 2005) (http://www.fia.fs.fed.us/tools-data/).
References
Smith, W.B., P.O. Miles, T.S. Vissage, and S.A. Pugh.
' J o ' o
2004. Forest resources of the United States, 2002. General
Technical Report NC-241. St. Paul, MN: USDA Forest
Service, North Central Research Station.
Smith, W.B., J.S. Vissage, D.R. Darr, and R.M. Sheffield.
2001. Forest resources of the United States, 1997. General
Technical Report NC-219. St. Paul, MN: USDA Forest
Service, North Central Research Station.
USDA Forest Service. 2005. Forest Inventory and Analysis,
national FIA data base systems. Accessed 2005.
94
-------
INDICATI
Forest Extent and Type
E>
CO
E
o
ro
c
_o
'E,
ro
o
ro
c
o O
O
^ HS 1=]
Reporting period
iverage: All 50 states.
ata source: Smith et a/., 2001, 2004
Exl
CO
E
o
ro
c
o
J_
ro
o
<
aCo
de
is
Da
libit
140
120
100
80
60
40
20
0
140
120
100
80
60
40
20
0
irerag
ined
imbe
fa soi
6-3. Timberland area in the eastern U.S. by forest type, 1953-20023
White- red-jack pine
Spruce-fir Longleaf-slash pine LobloHy-shortleaf pine 1 Oak-pin
H HUH nil
3 Oak-hickory
'53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02
Oak-gum-cypress
linn
1 Elm-ash-cottonwood Maple-beech-birch 1 Aspen-birch Non-stock
• II
'53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02 '53 '63 77 '87 '97 '02 '53 '63 77 '87
£- -^
97 '02
Year
B: States in the eastern U.S., based on USDA Forest Service reporting regions (see map at right). These data cover timberland, as
)y the Forest Service's Forest Inventory and Analysis (FIA) Program. Approximately 94% of the forest land in the eastern states
land.
irce: Smith etal., 2001, 2004
95
-------
INDICATO
Forest Extent and Type
Exhibit 6-4. Timberland area in the western U.S. by forest type, 1953-20023
Douglas-fir
Lodgepole pine
Dim
Ponderosa-Jeffrey Western white pine
pine
Fir-spruce
Larch
Redwood
Other western
softwood types
Western hardwood
types
Pinyon-juniper
'53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02
Non-stocked
'53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02 '53'63 77'87'97'02
Year
Coverage: States in the western U.S. (including Alaska and Hawaii), based on USDA Forest Service reporting
regions (see map at right). These data cover timberland, as defined by the Forest Service's Forest Inventory
and Analysis (FIA) Program. Approximately 39% of the forest land in the western states is timberland.
Data source: Smith etal., 2001, 2004
East
96
-------
INDICATO
Forest Fragmentation
The amount of forest land in the U.S. monitored by the
U.S. Department of Agriculture (USDA) Forest Service
has remained nearly constant over the past century, but the
patterns of human land use have affected its distribution
from one region of the U.S. to another. Forest fragmen-
tation involves both the extent of forest and its spatial
pattern, and is the degree to which forested areas are being
broken into smaller patches and pierced or interspersed
•with non-forest cover.
Forest fragmentation is a critical aspect of the extent and
distribution of ecological systems. Many forest species are
adapted to either edge or interior habitats. Changes in the
degree or patterns of fragmentation can affect habitat qual-
ity for the majority of mammal, reptile, bird, and amphib-
ian species found in forest habitats (Fahrig, 2003). As forest
fragmentation increases beyond the fragmentation caused
by natural disturbances, edge effects become more domi-
nant, interior-adapted species are more likely to disappear,
and edge- and open-field species are likely to increase.
This indicator of forest fragmentation was developed
by the USDA Forest Service. The indicator is based on the
2001 National Land Cover Database (NLCD), which was
constructed from satellite imagery showing the land area of
the contiguous U.S. during different seasons (i.e., leaves-on
and leaves-off) around the year 2001 (Homer et al., 2007).
The USDA Forest Service's Southern Research Station
performed a re-analysis of the NLCD, aggregating the four
NLCD forest cover classes (coniferous, deciduous, mixed,
and wetland forest) into one forest class and the remaining
land cover classes into a single non-forest class (USDA For-
est Service, 2007). A model that classifies forest fragmenta-
tion based on the degree of forest land surrounding each
forest pixel (a square approximately 30 meters on each
edge) for various landscape sizes (known as "windows")
provides a synoptic assessment of forest fragmentation for
the contiguous U.S. by assessing each pixel's "forest neigh-
borhood" within various distances.
Results are based on four degrees of forest cover: "core"
if a subject pixel is surrounded by a completely forested
landscape (no fragmentation), "interior" if a subject pixel is
surrounded by a landscape that is 90 to 100 percent forest,
"connected" if a subject pixel is surrounded by a landscape
that is 60 to 90 percent forest, and "patchy" if the subject
pixel is surrounded by less than 60 percent forest. The
window (landscape) size used for this analysis was 13 by
13 pixels, 390 meters on each edge, or about 15.2 hectares
(37.6 acres). The window is shifted one pixel at a time over
the map, so the target population for the indicator is all
forested pixels in the contiguous U.S. Percent forest was
resampled from 30-meter pixel data and aggregated by
state to develop the EPA Region-specific breakouts.
Exhibit 6-5. Forest fragmentation in the
contiguous U.S. by EPA Region, based on
2001 NLCDab
Degree of forest cover:0
Core
Interior
Connected
Patchy
38.0
26.7
27.8
7.5
33.5
33.3
23.5
23.6
28.7
14.3
30.3
22.1
21.4
23.1
35.9
22.8
33.8
12.8
19.0
22.0
Percent of forested pixels in each category:
Region 1
Region 2
Region 3
Region 4
Region 5
Region 6
Region 7
Region 8
Region 9
Region 10
All U.S.
Coverage: Areas of the contiguous
48 states classified as "forested" by
the 2001 National Land Cover
Database (NLCD).
bTotals may not add to 100% due to
rounding.
cSee text for definitions of forest
cover categories.
Data source: USDA Forest Service, 2007
23.0
15.6
21.0
15.4
32.3
31.0
23.7
38.0
27.8
22.8
29.2
20.2
29.7
22.5
29.4
18.4
29.4
26.0
31.9
12.8
26.1
22.9
32.1
18.9
EPA Regions
What the Data Show
Slightly more than 26 percent of the forested pixels in the
U.S. represent "core" forest, i.e., landscapes dominated
by forest (Exhibit 6-5). However, the data for "interior"
and "core" forests suggest that fragmentation is extensive,
•with few large areas of complete, unperforated forest cover.
About 19 percent of forest pixels in the U.S. occur in a
landscape where less than 60 percent of the "neighbor-
hood" is forest (i.e., forest cover is "patchy").
97
-------
Forest Fragmentation
There is considerable regional variation in forest frag-
mentation (Exhibit 6-5). Regions 1, 2, and 3 have more
than 30 percent "core" forest pixels, while fewer than 20
percent of the forest pixels in Region 7 are "core" forest.
From the opposite perspective, fewer than 10 percent of
forest pixels in Region 1 are surrounded by less than 60
percent forest, compared to almost 40 percent of the forest
pixels in Region 7.
Indicator Limitations
• Trend information is not available for this indicator.
Although earlier land cover data are available as part of the
1992 NLCD, they are not directly comparable with the
2001 NLCD due to differences in classification methodol-
ogy. Efforts to compare these two products are ongoing.
• The apparent degree of connectivity depends on the size
of the window. In a similar analysis of 1992 NLCD data,
Riitters (2003) determined that the percentages for all
categories (especially "core" and "connected" forest pixels)
decrease rapidly as the size of the window is increased
progressively from 18 to 162, 1,459, and 13,132 acres.
• Because the non-forest land cover classes were aggre-
gated, this indicator does not distinguish between natural
and anthropogenic fragmentation (although such a dis-
tinction has been made for global fragmentation by Wade
etal.,2003).
• The data do not include Hawaii or Alaska, which account
for about 1 out of every 6 acres of forest land in the U.S.
Data Sources
An earlier version of this analysis was published in Riitters
(2003) and Heinz Center (2005). The analysis presented
here has not yet been published; data were provided by the
USDA Forest Service (2007), and EPA grouped the results
by EPA Region. This indicator is based on land cover data
from the 2001 NLCD (MRLC Consortium, 2007).
References
Fahrig, L. 2003. Effects of habitat fragmentation on biodi-
versity. Annu. Rev. Ecol. Evol. Syst. 34:487-515.
Homer, C., J. Dewitz, J. Fry, M. Coan, N. Hossain, C.
Larson, N. Herold, A. McKerrow, J.N. VanDriel, andj.
Wickham. 2007. Completion of the 2001 National Land
Cover Database for the conterminous United States.
Photogramm. Eng. Rem. S. 73(4):337-341.
The H. John Heinz III Center for Science, Economics, and
the Environment. 2005. Forest pattern and fragmentation.
In: The state of the nation's ecosystems: Measuring the
lands, \vaters, and living resources of the United States. New
York, NY: Cambridge University Press. Web update 2005.
MRLC Consortium. 2007. National Land Cover Database
2001 (NLCD 2001). Accessed 2007.
Riitters, K.H. 2003. Report of the United States on the
criteria and indicators for the sustainable management of
temperate and boreal forests, criterion 1: Conservation of
biological diversity, indicator 5: Fragmentation of forest
types. Final report. FS-766A. In: Darr, D., ed. Data report:
A supplement to the National Report on Sustainable Forests.
Washington, DC: USDA Forest Service.
USDA Forest Service. 2007. Data provided to EPA by Kurt
Riitters, USDA Forest Service. September 18, 2007.
Wade, T.G., K.H. Riitters, J.D. Wickham, and K.B. Jones.
2003. Distribution and causes of global forest fragmentation.
Conserv. Ecol. 7(2):7.
98
-------
Fish Faunal Intactness
ntactness, the extent to which ecological communities
have retained their historical composition, is a critical
aspect of the biological balance of the nation's ecological
systems (NRC, 2000). It is of particular importance in
freshwater systems that are impacted by pollution, habitat
alteration, fisheries management, and invasive species.
This indicator tracks the intactness of the native fresh-
water fish fauna in each of the nation's major -watersheds
by comparing the current faunal composition of those
•watersheds with their historical composition. In this case,
historical data are based on surveys conducted prior to
1970. The indicator specifically measures the reduction
in native species diversity in each 6-digit U.S. Geological
Survey hydrologic unit code (HUC) cataloguing unit in
the 48 contiguous states. Intactness is expressed as a percent
based on the formula:
reduction in diversity — 1 —
# of current native species
\# of historical native species/
The native species diversity indicator proposed by the
National Research Council (NRC, 2000) compared expected
native species diversity (projected from species-area-curve
models) with observed diversity. This "Fish Faunal Intact-
ness" indicator makes use of empirical, rather than modeled,
data sets and focuses on a well-known group of organisms
•with a fairly strong historical record.
Reductions in -watershed diversity may be due either to
the overall extinction of a species (at least 12 U.S. fresh-wa-
ter fish species are known to be extinct and another three
species are known only from historical records and may be
extinct) or, more commonly, to the extirpation of a species
from selected-watersheds. In the case of regional extirpa-
tions, opportunities may exist for restoring a species to
•watersheds in its historical range.
The fish distributional data underlying this indica-
tor -were gathered by NatureServe, a nonprofit research
organization, and are derived from a number of sources,
including species occurrence data from state Natural Heri-
tage Programs, a broad array of relevant scientific literature
(e.g., fish faunas), and expert review in nearly every state.
These data -were assembled during the 1997-2003 period.
The underlying data include distributions for 782 native
fresh-water fish species across small -watersheds (8-digit
HUC). For this indicator, data -were pooled and reported
by larger 6-digit HUCs to reduce potential errors of omis-
sion in the smaller -watersheds.
What the Data Show
Watersheds covering about one-fifth (21 percent) of the
area of the contiguous U.S. appear to have fish faunas
that are fully intact, retaining the entire complement of
Exhibit 6-10. Percent reduction in native fish species diversity in the contiguous U.S. from
historical levels to 1997-20033
Percent of area
in each category
,7%
1-2%
21%
aData are displayed by 6-digit hydrologic unit code (HUC) watershed. Percent reduction is based on the number of
native species present during the period 1997-2003, compared with historical numbers documented prior to 1970. A
species is considered "present" if there is at least one record of its presence in any 8-digit HUC within the 6-digit HUC.
Data source: NatureServe, 2006
99
-------
Fish Faunal Intactness
Exhibit 6-11. Reduction in native fish species diversity in the contiguous U.S. from historical
levels to 1997-20033
Percent of area
in each category
aData are displayed by 6-digit hydrologic unit code (HUC) watershed. Reduction is based on the number of native
species present during the period 1997-2003, compared with historical numbers documented prior to 1970. A
species is considered "present" if there is at least one record of its presence in any 8-digit HUC within the 6-digit HUC.
Data source: NatureServe, 2006
fish species that were present before 1970 (Exhibit 6-10).
Watersheds covering nearly a quarter (24 percent) of the
area, however, have lost 10 percent or more of their native
fish species. Reductions in diversity are especially severe in
the South-west (e.g., the lower Colorado River -watershed)
and the Great Lakes, -with eight major watersheds (repre-
senting 2 percent of total area) having lost at least half of
their native fish species.
Some -watersheds are naturally more species-rich than
others, and for those -with greater historical diversity,
even a small percentage reduction may mean the loss of
numerous species in absolute terms. Although the great-
est diversity offish species is found in the Southeast, the
greatest reduction in numbers has occurred in portions of
the Mid-west and the Great Lakes, -where several -watersheds
have lost more than 20 species (Exhibit 6-11). In contrast,
south-western HUCs have all lost 10 or fewer species, but
because these -watersheds historically supported fewer spe-
cies, on a percentage basis their fish faunas are regarded as
less intact.
Indicator Limitations
• The incomplete historical record for fresh-water fish dis-
tributions and inconsistent inventory records for contem-
porary fish distributions are sources of uncertainty.
• Although NatureServe has attempted to compile the
most complete distributional information possible for
these species at the 8-digit HUC level, these data are
dynamic; new records frequently are added and existing
records are revised as new information is received and as
taxonomic changes occur.
Data Sources
This indicator presents a summary of data available from the
NatureServe Explorer database (NatureServe, 2006) (http://
www.natureserve.org/getData/dataSets/watershedHucs/
index.jsp). The identity and status (current vs. historical)
of all native fish species recorded in each 8-digit HUC are
available from this database, along -with species-by-species
distribution maps at the 8-digit HUC level. Analyses based
on these data have previously been reported in Master et al.
(1998, 2003) and Stem et al. (2000).
References
Master, L., A. Olivero, P. Hernandez, and M. Anderson.
2003. Using small -watershed fish, mussel, and crayfish
historical and current presence data to describe aquatic bio-
geography and inform its conservation. Abstract #PO67.
Society for Conservation Biology Annual Meeting,
Duluth, Minnesota.
Master, L.L, S.R Flack, andB.A. Stem. 1998. Rivers of life:
Critical -watersheds for protecting fresh-water biodiversity.
Arlington, VA: The Nature Conservancy.
100
-------
Fish Faunal Intactness
NatureServe. 2006. NatureServe explorer. Accessed 2006.
NRC (National Research Council). 2000. Ecological
indicators for the nation. Washington, DC: National
Academies Press.
Stem, B.A., L.S. Kutner, andJ.S. Adams. 2000. Precious
heritage: The status of biodiversity in the United States.
New York, NY: Oxford University Press.
101
-------
INDICATO
Non-Indigenous Benthic Species in the Estuaries of the
Pacific Northwest
Non-indigenous species (NIS) are one of the greatest
threats to aquatic ecosystems and can impact local
and regional economies (Lowe et al., 2000). The number
of invasive species in estuaries of the Pacific North-west
(including Puget Sound, Columbia Estuary, and Coos Bay)
is rising, and these areas can become sources of invasives to
other locales. Coastal waters are particularly vulnerable to
NIS transported in ballast water and introduced via aqua-
culture (Puget Sound Action Team, 2002). It is becoming
apparent that NIS are capable of impacting estuaries along
the Pacific coast, even though they are rarely addressed in
routine monitoring studies. One limitation is the lack of
standardized invasion metrics and threshold values.
This indicator focuses on estuarine soft-bottom com-
munities of the Columbian Biogeographic Province located
along the Pacific coast from Cape Mendocino, California,
north to the Strait of Juan de Fuca at the entrance to Puget
Sound, Washington. It is limited to sites with salinities of 5
parts per thousand or higher. The indicator is based on the
percent abundance of NIS individuals relative to the com-
bined abundance of native and NIS individuals in a benthic
grab sample.
The data for this indicator were collected by EPA's
Environmental Monitoring and Assessment Program
(EMAP) using a probability survey over the 1999-2001
period (Nelson et al., 2004, 2005) and by a special proba-
bilistic study focusing on estuaries not exposed to ballast
•water or aquaculture. Probability sampling provides unbi-
ased estimates of the percent abundance of natives and NIS
in all estuaries in the study area, but because the data for
the special study have not yet been statistically expanded,
data for this indicator are based on stations sampled rather
than area.
Interpretation of this indicator requires threshold val-
ues to distinguish among different levels of invasion. To
determine the lowest expected level of invasion within the
Columbian Biogeographic Province, EPA examined the
extent of invasion in estuaries with minimal exposure to
ballast \vater discharges and aquaculture of exotic oysters,
•which are the primary invasion vectors in the region.
Using observed percentages of NIS at the minimally
exposed estuaries as a reference, the threshold for "mini-
mally invaded" survey sites was set at 10 percent NIS (i.e.,
sites \vere classified as minimally invaded if NIS consti-
tuted 0 to 10 percent of the individuals collected). Survey
sites \vere classified as "highly invaded" if NIS were more
abundant than native species (more than 50 percent NIS)
and as "moderately invaded" if NIS constituted 10 to 50
percent of the individuals.
Exhibit 6-12. Relative abundance of
non-indigenous benthic species in estuaries of
the Pacific Northwest, 1999-2001ab
Extent of invasion:
Minimal0
Moderate11
High6
65.7
19.9
14.5
Percent of estuarine sites in each category:
All
estuaries
Exposed
estuaries'
Minimally
exposed
estuaries'
56.1
28.6
15.3
79.4
7.4
13.2
Study
area
Coverage: Soft-bottom estuaries
between Cape Mendocino, CA, and
the Strait of Juan de Fuca, WA
(limited to sites with salinity ^5
parts per thousand).
bTotals may not add to 100% due to
rounding.
"Minimally invaded: 0-10%
of benthic organisms belong to
non-indigenous species
dModerately invaded: >10-50% of
benthic organisms belong to non-indigenous species
eHighly invaded: >50% of benthic organisms belong to
non-indigenous species
'"Exposed" estuaries have been exposed to ballast water
discharges from international shipping and/or aquaculture of
exotic oysters. "Minimally exposed" estuaries have not.
Data source: U.S. EPA, 2006
What the Data Show
Approximately 15 percent of the stations in the Columbian
Province were highly invaded (i.e., abundance of NIS was
greater than abundance of natives) and another 20 per-
cent \vere moderately invaded (Exhibit 6-12). The EMAP
survey showed that NIS were among the most frequently
occurring anthropogenic stressors in this biogeographic
region when compared to indicators of sediment contami-
nation or eutrophication (Nelson et al., 2004).
The extent of invasion was not uniform, however,
among exposed and minimally exposed estuaries. Estuaries
•with greater exposure to these invasion vectors were more
invaded; 44 percent of the stations in the exposed estuaries
102
-------
INDICATO
Non-Indigenous Benthic Species in the Estuaries of the
Pacific Northwest
•were moderately to highly invaded compared to only 21
percent of the stations in minimally exposed estuaries
(Exhibit 6-12). Nonetheless, the observation that 21 per-
cent of the stations in these "pristine" estuaries were at least
moderately invaded indicates that NIS can disperse widely
once they are introduced into a region, so even estuaries
•with no direct exposure to ballast water or aquaculture are
at risk of invasion.
Indicator Limitations
• This indicator presents baseline data only; trend
information is not yet available.
• Studies in the San Francisco Estuary (Lee et al., 2003)
and in Willapa Bay, Washington (Ferraro and Cole,
in progress) have shown that the percent of NIS can
vary substantially among different types of soft-bottom
communities—e.g., unvegetated sediment versus sea
grass beds. Thus, regional background values for the
Columbian Province as a whole may not be appropriate
for specific community types.
• This indicator represents percent NIS in individual ben-
thic grabs of the soft-bottom community, but does not
characterize the total number of NIS in the estuaries. It
does not include benthic NIS not subject to grab sam-
pling, particularly hard substrate organisms.
• The data for the indicator were only collected during
a summer index period and thus do not capture
seasonal variations.
• The threshold values for "minimally invaded," "moder-
ately invaded," and "highly invaded" are preliminary
and require further research in order to establish their
ecological significance. Specific values may differ in
other biogeographic provinces.
Data Sources
Data for this indicator were collected by two different
studies: EPA's National Coastal Assessment (NCA) and a
special EPA study of minimally exposed estuaries. The
complete results from these studies were not publicly avail-
able at the time this report went to press, but summary data
from the 1999 NCA are available from Nelson et al. (2004,
2005), and the underlying sampling data can be obtained
from EPA's NCA database (U.S. EPA, 2007) (http://
www.epa.gov/emap/nca/html/data/index.html). Results
from the special study of minimally exposed estuaries will
be published in the near future. Until then, data for this
indicator can be obtained from EPA's Western Ecology
Division (U.S. EPA, 2006).
References
Lee II, H., B. Thompson, and S. Lowe. 2003. Estuarine
and scalar patterns of invasion in the soft-bottom benthic
communities of the San Francisco Estuary. Biol. Invasions
5:85-102.
Lowe, S., M. Browne, S. Boudjelas, and M. De Poorter.
2000. 100 of the -world's \vorst invasive alien species—a
selection from the Global Invasive Species Database. Auck-
land, New Zealand: International Union for the Conserva-
tion of Nature, Invasive Species Specialist Group.
Nelson, W.G., H. Lee II, andj. Lamberson. 2005. Condi-
tion of estuaries of California for 1999: A statistical sum-
mary. EPA/620/R-05/004.
Nelson, W.G., H. Lee II, J.O. Lamberson, V. Engle, L.
Harwell, and L.M. Smith. 2004. Condition of estuaries of
the -western United States for 1999: A statistical summary.
EPA/620/R-04/200.
Puget Sound Action Team. 2002. Puget Sound update:
The eighth report of the Puget Sound Ambient Moni-
toring Program (PSAMP). Olympia, WA: Puget Sound
Water Quality Action Team,
U.S. EPA (United States Environmental Protection Agency).
2007. National Coastal Assessment. Accessed 2007.
U.S. EPA. 2006. Data provided to ERG (an EPA
contractor) by Henry Lee, EPA Western Ecology Division.
August 7, 2006.
103
-------
INDICATOR
Carbon Storage in Forests
After carbon dioxide is converted into organic matter by
photosynthesis, carbon is stored in forests for a period
of time in a variety of forms before it is ultimately returned
to the atmosphere through the respiration and decomposi-
tion of plants and animals, or harvested from forests for use
in paper and wood products. A substantial pool of carbon
is stored in woody biomass (roots, trunks, and branches).
Another portion eventually ends up as organic matter in
forest floor litter and the upper soil horizons. Carbon stor-
age in forest biomass and forest soils is an essential physical
and chemical attribute of stable forest ecosystems, and a
key link in the global carbon cycle.
This indicator, developed by the U.S. Department of
Agriculture (USDA) Forest Service, tracks decadal changes
in net carbon storage rates in the pools of living and dead
biomass in forests in the contiguous 48 states. The carbon
pools for this indicator are estimated using USDA Forest
Service Forest Inventory and Analysis (FIA) data from five
historical periods (circa 1953, 1963, 1977, 1987, and 1997).
These data cover forest classified as "timberland" under
FIA data collection procedures—that is, forests capable of
producing at least 20 cubic feet per acre per year of industrial
•wood and not withdrawn from timber utilization by statute
or regulation. Timberland makes up roughly two-thirds of
U.S. forest land. Alaska and Hawaii are not included because
of limited historical data. The FIA program estimates
carbon storage using on-the-ground measurements of tree
trunk size from many forest sites; statistical models that
show the relationship between trunk size and the weight of
branches, leaves, coarse roots (greater than 0.1 inch in diam-
eter), and forest floor litter; and estimates of forest land area
obtained from aerial photographs and satellite imagery. Val-
ues are converted into carbon storage based on coefficients
derived from previous field studies (Smith and Heath, 2002;
Smith et al, 2003; Birdsey, 1996). Forest floor litter is com-
posed of dead organic matter above the mineral soil hori-
zons, including litter, humus, and fine woody debris. Larger
branches and logs on the ground are counted as "down dead
•wood." Organic carbon in soil is not included.
What the Data Show
The change in carbon inventories from year to year—
i.e., net storage—reflects increases in growth as well as
decreases due to harvesting, land use change, and dis-
turbances such as fire, insects, and disease. Overall, net
carbon storage in forests of the contiguous 48 states has
been positive since 1953 (Exhibit 6-13), indicating that
over at least the last half-century, forests have served as a
sink rather than a source of carbon. The average rate of net
carbon storage in forests increased between the 1950s and
the 1980s, peaking at 210 million metric tons of carbon
per year (MtC/yr) from 1977 to 1986. The rate declined to
Exhibit
storage
6-13. Average annual net carbon
in forests of the contiguous U.S
.,by
forest component, 1 953-1 996a
CD
O)
o _ 200
HI 150
C CD
15 E 100
1|
§> E 50
i
1953-1962 1963-1976
1977-1986
1987-1996
Reporting period
"Coverage: Forest land
classified as "timberland,"
which accounts for
approximately two-thirds
of the forest land of the
contiguous 48 states.
These data do not include
carbon stored in forest soi
Data source: USDA Forest
Service, 2004a,b
D Aboveground live trees
DAboveground standing
dead trees
D Understory vegetation
D Down dead wood
(including stumps)
D Forest floor litter
D Belowground live trees (roots)
• Belowground dead wood
135 MtC/yr for the last period of record (1987-1996), with
declining storage evident in live, dead, and understory
pools. This decline is thought to be due to a combination
of increased harvests relative to growth, more accurate
data, and better accounting of emissions from dead wood
(USDA Forest Service, 2004b). The rate of storage over
this period is equivalent to approximately 9 to 10 percent
of U.S. carbon dioxide emissions over a comparable period
(U.S. EPA, 2005).
Carbon storage trends vary among regions of the
country, depending on land use patterns and factors such as
climate and soil quality. In three of the four major regions,
net storage was positive throughout the period of record,
•with the North generally showing the largest net storage
rates (Exhibit 6-14). The exception was the Pacific Coast
region, which experienced net losses of forest carbon dur-
ing two of the four reporting periods. Rates of net carbon
storage appear to have decreased over time in the South;
this trend is thought to be due to an increase in harvesting
relative to growth (USDA Forest Service, 2004b). Some of
the harvested carbon is sequestered in wood products.
104
-------
INDICATOR
Carbon Storage in Forests
Exhibit 6-14. Average annual net carbon storage in forests of
the contiguous U.S. by region, 1953-1996a
100
50
25
-25
i North
i South
Rocky
Mountains
i Pacific
Coast
1953-1962 1963-1976 1977-1986
Reporting period
Coverage: Forest land classified as
"timberland," which accounts for
approximately two-thirds of the forest land
of the contiguous 48 states. These data do
not include carbon stored in forest soil.
Data source: USDA Forest Service,
2004a,b
1987-1996
Rocky
Mountains
Indicator Limitations
• The data include only forest classified as "timberland,"
•which excludes about one-third of U.S. forest land cover.
Historical data from Alaska and Hawaii are insufficient
for inclusion in this indicator.
• Data are derived from state inventories that do not cor-
respond exactly to the years identified in Exhibits 6-13
and 6-14.
• Carbon stored in forest soil is not included.
• Carbon pools are not measured, but are estimated based
on inventory-to-carbon coefficients developed with infor-
mation from ecological studies. These coefficients may
change over time as new ecological studies are conducted,
•which could change storage rate estimates.
These limitations are discussed in detail in Heath and
Smith (2000) and Smith and Heath (2000, 2001).
Data Sources
Exhibits 6-13 and 6-14 were previously published in the data
supplement to USDA Forest Service (2004b). The numbers
depicted in these figures have not been published, but were
provided by the USDA Forest Service (2004a). The physical
measurements used as inputs in the carbon storage models
can be obtained from the FIA database (USDA Forest Ser-
vice, 2005) (http://fia.fs.fed.us/tools-data/).
References
Birdsey, R. A. 1996. Carbon storage for major
forest types and regions in the conterminous
United States. In: Sampson, R.N., and D.
Hair, eds. Forests and global change, volume
2: Forest management opportunities for miti-
gating carbon emissions. Washington, DC:
American Forests, pp. 1-25, 261-308.
Heath, L.S., and J.E. Smith. 2000. An assess-
ment of uncertainty in forest carbon budget
projections. Environ. Sci. Policy 3:73-82.
Smith, J.E., andL.S. Heath. 2002. Estimators
of forest floor carbon for United States forests.
Res. Pap. NE-722. Newtown Square, PA:
USDA Forest Service, Northeastern Research
Station. 37 pp.
Smith,J.E., andL.S. Heath. 2001. Identify-
ing influences on model uncertainty: An
application using a forest carbon budget
model. Environ. Manage. 27:253-267.
Smith,J.E., andL.S. Heath. 2000. Con-
siderations for interpreting probabilistic
estimates of uncertainty of forest carbon. In:
Joyce, L.A., and R. Birdsey, eds. The impact
of climate change on America's forests.
General Technical Report RMRS-59. Fort Collins, CO:
USDA Forest Service, Rocky Mountain Research Station.
pp. 102-111.
Smith, J.E., L.S. Heath, and J.C. Jenkins. 2003. Forest
volume-to-biomass models and estimates of mass for live
and standing dead trees of U.S. forests. General Techni-
cal Report NE-298. Newtown Square, PA: USDA Forest
Service, Northeastern Research Station. 57 pp.
USDA Forest Service. 2005. Forest Inventory and Analysis
(FIA) database. Accessed 2005.
USDA Forest Service. 2004a. Data provided to ERG (an
EPA contractor) by Linda Heath, USDA Forest Service.
December 23, 2004.
USDA Forest Service. 2004b. National report on sus-
tainable forests—2003. (main site); (data supple-
ment: summary); (data supplement:
graphics and metadata)
U.S. EPA (United States Environmental Protection Agency).
2005. Inventory of U.S. greenhouse gas emissions and sinks:
1990-2003. EPA/430/R-05/003.
105
-------
INDICATOR
U.S. and Global Mean Temperature and Precipitation
Air temperature and precipitation are two important
properties of climate and are the most widely measured
variables. Changes in these indicators may have -wide-
ranging direct or indirect effects on ecological condition
and human health. These impacts may be positive or nega-
tive, depending on the effect, the magnitude of change,
and the location. For example, changes in temperature can
affect heat- and cold-related mortality and illness due to
altered frequency and magnitude of heat waves and cold
spells. Changes in temperature may also change the range
and distribution of animal and plant species. Precipitation
changes affect water availability and quality, which can
have important effects on agricultural, forest, animal, and
fisheries productivity, as well as human nutrition. Indirect
effects of temperature and precipitation changes include
changes in the potential transmission of vector-borne
infectious diseases. These may result from alterations in the
ranges and seasons of animals that carry disease or from
accelerated maturation of certain infectious parasites.
This indicator shows trends in temperature and precipi-
tation based on instrumental records from 1901 to 2006
(except for Alaska and Hawaii, where records begin in
1918 and 1905, respectively). Air temperature and precipi-
tation trends are summarized for the contiguous U.S., as
•well as for 11 climate regions of the U.S., including Alaska
and Hawaii (these climate regions are different from the
ten EPA Regions). For context, this indicator also shows
trends in global temperature (over land and sea) and global
precipitation (overland) from 1901 to 2006.
Temperature and precipitation data are presented as
trends in anomalies. An anomaly represents the difference
between an observed value and the corresponding value
from a baseline period. This indicator uses a 30-year base-
line period of 1961 to 1990. To generate the temperature
time series, measurements were converted into monthly
anomalies, in degrees Fahrenheit. The monthly anomalies
then \vere averaged to get an annual temperature anomaly
for each year. Precipitation trends were calculated in
similar fashion, starting with anomalies for total monthly
precipitation, in millimeters. Monthly anomalies were
added to get an annual anomaly for each year, which was
then converted to a percent anomaly—i.e., the percent
departure from the average annual precipitation during the
baseline period. Trends in temperature and precipitation
•were calculated from the annual time series by ordinary
least-squares regression. For each of the 11 climate regions,
this indicator also shows a smoothed time series, which was
created from the annual series using a nine-point bino-
mial filter (4 years on each side, averaged with decreasing
•weights further from the center year).
What the Data Show
Since 1901, temperatures have risen across the contigu-
ous U.S. at an average rate of 0.12°F per decade (1.2°F per
Exhi
the c
3
2
LJT 1
"re o
o
< -1
-2
-3
19
3
2
CT 1
>1
B °
O
< -1
-2
-3
19
aAnom
Datas
bit 6-15. Annual temperature anomalies in
ontiguous U.S. and worldwide, 1901-20063
A. Contiguous U.S. temperature anomalies
1901-2006 trend: +1.17°F per century
1977-2006 trend: +5.92°F per century
> li.Jl.liiL .iiLlii
fp**^inT
00 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year
B. Global temperature anomalies
1901-2006 trend: +1.16°F per century
1977-2006 trend: +3.05°F per century
wppp.-^r>.^r*a
00 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year
jlies are calculated with respect to the 1961-1990 mean.
ource: NOM, 2007b
century) (Exhibit 6-15, panel A). Over the past 30 years,
average temperatures rose at an increased rate of 0.59°F
per decade, and 5 of the top 10 warmest years on record
for the contiguous U.S. have occurred since 1990. The
overall warming trend is not confined to just a few anoma-
lous years, as the last eight 5-year periods (2002-2006,
2001-2005, ...1995-1999) were the eight warmest 5-year
periods on record (NOAA, 2007a). Warming occurred
throughout the U.S., with all but three of the 11 climate
regions (all but the Central, South, and Southeast) show-
ing an increase of more than 1°F since 1901 (Exhibit 6-16).
The greatest temperature increase occurred in Alaska
(3.3°F per century).
Trends in global temperature and precipitation provide a
context for interpreting trends in temperature and precipita-
tion in the U.S. Instrumental records from land stations and
ships indicate that global mean surface temperature rose by
about 1.2°F during the 20th century (Exhibit 6-15, panel B),
106
-------
INDICATOR
U.S. and Global Mean Temperature and Precipitation
Exhibit 6-16. Annual temperature anomalies in the U.S. by region, 1901-20063
A. Northeast
4 1901-2006 trend: +1.79°F per century
3
2
1
0
-1
-2
-3
-4
-5
1900 1920 1940 1960 1980 2000
D. South
4 1901-2006 trend: +0.17T per century
3
2
1
0
-1
-2
-3
-4
-5
1900 1920 1940 1960 1980 2000
G. Southwest
4 1901-2006 trend: +1.69T per century
3
2
1
0
-1
-2
-3
-4
-5
1900 1920 1940 1960 1980 2000
J. Alaska
5 1918-2006 trend: +3.25°F per century
4
3
2
1
0
-1
-2
-3
-4
-5
1900 1920 1940 1960 1980 2000
B. Southeast
C. Central
1901-2006 trend: no change
'
-4
-5 I
1900 1920 1940 1960 1980 2000
E. East North Central
4 1901-2006 trend: +0.24T per century
3
2
1
0
-1
-2
-3
-4
-5
1900 1920 1940 1960 1980 2000
F. West North Central
1901-2006 trend:
+1.76°F per century
1901-2006 trend: +1.81 °F per century
1901-2006 trend:+2.11°F per century
1901-2006 trend: +1.73°F per century
-5
1900 1920 1940 1960 1980 2000
K. Hawaii
1900 1920 1940 1960 1980 2000
1905-2006 trend: +1.15°F per century
5
4
3
2
1
0
-1
-2
-3
-4
-5
1900 1920 1940 1960 1980 2000
Northwest
West
North Central Central
East
North Central
Year
Anomalies are calculated with respect to the 1961-1990 mean.
bTime series were smoothed using a 9-point binomial filter.
Data source: NOAA, 2007b
— Annual anomaly
— Smoothed trendb
Northeast
West 7
Southwest
Temperature change (°F per century):
-3-2-1012
Gray interval:-0.1 to0.1°F
107
-------
INDICATOR
U.S. and Global Mean Temperature and Precipitation
similar to the rate of warming within the contiguous U.S.
During the last three decades, however, the U.S. warmed at
nearly twice the global rate.
As global mean temperatures have risen, global mean
precipitation also has increased (Exhibit 6-17, panel B).
This is expected because evaporation increases with
increasing temperature, and there must be an increase in
precipitation to balance the enhanced evaporation (IPCC,
2007). Globally, precipitation over land increased at a
rate of 1.7 percent per century since 1901, but the trends
vary spatially and temporally. Over the contiguous U.S.,
total annual precipitation increased at an average rate of
6.5 percent per century since 1901 (Exhibit 6-17, panel
A), although there was considerable regional variability
(Exhibit 6-18). The greatest increases came in the East
North Central climate region (11.2 percent per century)
and the South (10.5 percent). Hawaii was the only region
to show a decrease (-7.2 percent).
Indicator Limitations
• Biases may have occurred as a result of changes over time
in instrumentation, measuring procedures (e.g., time of
day), and the exposure and location of the instruments.
Where possible, data have been adjusted to account for
changes in these variables.
• Uncertainties in both the temperature and precipitation
data increase as one goes back in time, as there are fewer
stations early in the record. However, these uncertainties
are not sufficient to mislead the user about fundamental
trends in the data.
Data Sources
Anomaly data were provided by the National Oceanic and
Atmospheric Administration's (NOAA's) National Climatic
Data Center (NCDC), which calculated global, U.S., and
regional temperature and precipitation time series based
on monthly values from a net-work of long-term monitor-
ing stations (NOAA, 2007b). Data from individual stations
•were obtained from the U.S. Historical Climate Net-work
(USHCN version 1) and the Global Historical Climate
Network (GHCN), which are NCDC's online databases
(NOAA, 2007c).
References
IPCC (Intergovernmental Panel on Climate Change).
2007. Climate change 2007: The physical science basis.
Contribution of Working Group I to the fourth assessment
report of the Intergovernmental Panel on Climate Change.
Cambridge, UK: Cambridge University Press.
Exhibit 6-17. Annual precipitation anomalies in
the contiguous U.S. and worldwide, 1901-20063
20
15
>• 10
ro
I 5
™ o
1 -5
Q_
-10
-15
-20
A. Contiguous U.S. precipitation anomalies
1901-2006 trend: +6.46% per century
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year
20
15
B. Global precipitation anomalies
>• 10
ro
g 5
1 901 -2006 trend: +1 .74% per century
I*"''!
Kv*-
U. t J^J
1 -5
o.
-10
-15
-20
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year
Anomalies and percent change are calculated with respect to the
1961-1990 mean.
Data source: NOAA, 2007b
NOAA. 2007a. 2006 annual climate review: U.S. sum-
mary. June 21, 2007, edition,
NOAA. 2007b. Data provided to ERG (an EPA contrac-
tor) by Jay Lawrimore and David Wuertz, NOAA. October
12-November 16, 2007.
NOAA. 2007c. National Climatic Data Center. Accessed
October-November 2007. (NCDC home page); (U.S.
Historical Climate Net-work version 1);
(Global Historical Climate Net-work)
108
-------
INDICATOR
U.S. and Global Mean Temperature and Precipitation
Exhibit 6-18. Annual precipitation anomalies in the U.S. by region, 1901-20063
100
80
60
40
20
0
-20
-40
-60
A. Northeast
1901-2006 trend: +8.26% per century
1900 1920 1940 1960 1980 2000
100
80
60
40
20
0
-20
-40
-60
D. South
1901-2006 trend: +10.51% per century
1900 1920 1940 1960 1980 2000
100
80
60
40
20
0
-20
-40
-60
G. Southwest
1901-2006 trend: +1.28% per century
1900 1920 1940 1960 1980 2000
100
80
60
40
20
0
-20
-40
-60
J. Alaska
1918-2006 trend: +5.92% per century
1900 1920 1940 1960 1980 2000
100
B. Southeast
1901-2006 trend:+1.87% per century
1900 1920 1940 1960 1980 2000
100
80
60
40
20
0
-20
-40
-60
E. East North Central
1901-2006 trend: +11.23% per century
1900 1920 1940 1960 1980 2000
100
80
60
40
20
0
-20
-40
-60
H. West
1901-2006 trend:
+9.13% per century!
1900 1920 1940 1960 1980 2000
100
80
60
40
20
0
-20
-40
-60
K. Hawaii
1905-2006 trend: -7.16% per century
1900 1920 1940 1960 1980 2000
Year
Anomalies and percent change are calculated with respect to the
1961-1990 mean.
bTime series were smoothed using a 9-point binomial filter.
Data source: NOM, 2007b
— Annual anomaly
—Smoothed trendb
C. Central
1901-2006 trend: +8.13% per century
100
80
60
40
20
0
-20
-40
-60
1900 1920 1940 1960 1980 2000
100
80
60
40
20
0
-20
-40
F. West North Central
1901-2006 trend:+1.91 % per century
1900 1920 1940 1960 1980 2000
100
I. Northwest
1901-2006 trend: +5.97% per century
1900 1920 1940 1960 1980 2000
Northwest
West
North Central Central
East /
North Central / Northeast
West
Southwest
Change in precipitation (% per century):
P
-35-28-21-14 -7 0 7 14 21 28 35
Gray interval: -2 to 2%
109
-------
Sea Level
Sea level is an indicator of global and local change and
a factor that affects human welfare and coastal ecosys-
tem conditions. Coastal areas host a rich set of natural and
economic resources and include some of the most developed
and rapidly growing population centers in the nation. More
than 100 million people globally live within 1 meter of the
mean sea level and more than 40 percent of the U.S. popula-
tion lives in -watersheds along U.S. ocean coasts (NOAA,
2005). Changing sea levels can inundate low-lying wetlands
and dry lands (Burkett et al., 2005), erode beaches (USGS,
1998), change rates of sedimentation (Olffet al., 1997),
and increase the salinity of marshes, estuaries, and aquifers
(Condrey et al., 1995; Williams et al., 1999). Documented
consequences of sea level rise include loss of buffering
against storms and floods (Burkett et al., 2005), changes in
bird populations (Erwin, 2005) and land cover (Williams et
al., 1999), property losses (Burkett et al., 2005), and infra-
structure damage (Theiler and Hammar-Klose, 1999; U.S.
Department of Transportation, 2003).
Approximately 58,000 square kilometers of land in the
contiguous U.S. lie less than 1.5 meters above sea level;
80 percent of this land is in Louisiana, Florida, Texas, and
North Carolina (Titus and Richman, 2001). Almost half
of the shoreline studied along the U.S. Atlantic Coast was
determined to be highly to very highly vulnerable to effects
of sea level rise (Theiler and Hammar-Klose, 1999). The
areas of highest vulnerability are high-energy coastlines
•where the coastal slope is low and the major landform type
is a barrier island. The risks may be minimal if wetlands
accretion can match or outpace sea level rises, but accretion
rates vary -widely (Hartig et al., 2000, Table 3).
A number of factors affect sea level, including, but not
limited to, changes in sea temperature, salinity, and total
•water volume and mass (e.g., from melting glaciers or
changes in the amount of water stored on land). Sea level
rises with warming sea temperatures and falls with cooling.
Changes in the total volume and mass of ocean water also
result from the melting or accumulation of Antarctic and
Greenland ice sheets and non-polar glaciers and changes
in the amount of water stored in lakes, rivers, and ground
•water. As such, global average sea level change is
an indicator of the physical and climatic stability of the
global environment.
Exhibit 6-20. Changes in relative sea level along U.S. coasts, 1950-1999a
75°N
60°N -
45°N -
30°N -
15°N
o
o
180°
165°W
150°W
135°W
120°W
105°W
90°W
75°W
60°W
aTrends are based on tidal gauge measurements. Each dot represents a tidal gauge
station that operated during the period 1950-1999.
Data source: NOAA, 2006
Mean relative sea level change (mm per year):
o-18to-15 o -5.99to-3 o 3.01 to 6
o -14.99to-12 o -2.99 to 0 o 6.01 to 9
o -11.99to-9 o 0.01(03 o 9.01 to 12
• -8.99to-6
110
-------
Sea Level
Temporal scale is an important factor in interpreting sea
level trends. Sea level changes may reflect factors such as
seasonality, inter-annual to decadal scale variability such
as El Nino, and/or long-term climate change (decades to
centuries). Spatial scale also is important because absolute
sea height does not change uniformly around the globe.
This indicator presents trends in absolute and relative
sea level. Absolute sea level represents only the sea height,
\vhereas relative sea level change is defined as sea height
change plus land height changes (due to subsidence or
uplift and changes in natural land accretion). Relative sea
level data are from the tidal gauge measurements of the
National Water Level Observation Net-work, composed
of approximately 175 long-term, continuously operat-
ing stations located along the U.S. coast, including the
Great Lakes and islands in the Atlantic and Pacific Oceans
(Smith, 1980; Gill and Schultz, 2001). Tidal gauge data
are presented from 1950 to 1999, although a few loca-
tions have been monitoring since the mid-1800s (NOAA,
2001). Absolute sea level data are from satellite measure-
ments from NASA's TOPEX/Poseidon spacecraft, which
uses radar to map the precise features of the ocean surface,
and the "Jason" satellite, which monitors ocean circulation
(Leuliette et al., 2006). The two satellites use radar altim-
etry to collect sea level data globally. These data have been
available since 1993.
What the Data Show
Relative sea levels (combined land and sea movement) in
many locations rose from 1950 to 1999, typically at rates
of 0-3 millimeters per year (mm/yr) (up to 1 foot per
century) (Exhibit 6-20). Relative sea level has risen more
rapidly (3-6 mm/yr) along the mid-Atlantic coast from
North Carolina to New Jersey and at rates as high as 9-12
mm/yr at two stations in Louisiana. Other locations, such
as the southern coast of Alaska, show relative sea level drop,
•with a maximum decrease of 16 mm/yr. Average relative
sea level rise for all U.S. coasts was not calculated because
the distribution of tidal gauge stations is not spatially repre-
sentative of aggregate trends, but for reference, an analysis
of tidal gauge data worldwide estimated that on average,
relative sea level rose between 1.5 and 2.0 mm/yr during
the 20th century (Miller and Douglas, 2004).
The satellite record shows that global mean absolute sea
level (i.e., independent of land movements) has increased
at a rate of 3 mm (0.12 inches) per year since 1993 (Exhibit
6-21). Absolute sea levels do not change uniformly around
the Earth, however. Around the U.S., areas with increas-
ing absolute sea level include the Gulf coast and portions of
the Atlantic coast (Exhibit 6-22). Areas showing a decrease
include the southern part of the Pacific coast and the -west-
ern Gulf of Alaska.
Exhibit 6-21. Global mean sea level, 1993-20063
1992 1994 1996 1998 2000 2002 2004 2006
Year
TOPEX measurements
Jason measurements
60-day smoothing
aValues are reported as anomalies with
respect to the 1993-1997 mean.
bData were collected by the
TOPEX/Poseidon and Jason 1 satellite
altimeters. Data were adjusted by
applying an inverse barometer (air pressure) correction and removing
seasonal signals.
Data source: Leuliette et al., 2006
Indicator Limitations
• An estimated 50 to 60 years of data are required to
obtain linear mean sea level trends having a 1 mm/yr
precision with a 95 percent statistical confidence interval.
• Tidal gauge measurements do not represent more gener-
alized (i.e., average) relative sea level change along U.S.
coasts (or globally).
• Most local tidal gauge measurements cannot indicate
•whether changes in relative sea level are due to changes
in absolute sea level or changes in land elevation.
• Satellite data are not available for a multi-decadal time
series needed to separate out medium-term variability
from long-term change.
• Satellite data are not horizontally precise enough to
resolve sea level trends for small water bodies (such as
many estuaries) or for localized interests (such as a par-
ticular harbor or beach).
Data Sources
Exhibit 6-20 is based on a map and corresponding trend
data published by the National Oceanic and Atmospheric
111
-------
Sea Level
Exhibit 6-22. Changes in absolute sea level along U.S. coasts, 1993-20063
75°N
60°N -
45°N -
30°N -
15°N
180°
165°W
150°W
135°W
120°W
105°W
75 °W
60°W
aTrends are based on satellite measurements. Data were adjusted by
applying an inverse barometer (air pressure) correction.
Data source: Leuliette et al., 2006
Mean absolute sea level change (mm per year):
No data -15-10 -5 0
10 15
Administration's (NOAA's) National Oceans Service
(NOAA, 2006) (http://tidesandcurrents.noaa.gov/sltrends/
sltrends.shtml). These data were previously published in
NOAA (2001), along with a list of station coordinates
(NOAA, 2001, Appendix I). Individual station measure-
ments are accessible through NOAA (2006).
Exhibits 6-21 and 6-22 were produced using data pro-
vided by Leuliette et al. (2006) (time series at http://sealevel.
colorado.edu/results.php; map at http://sealevel.colorado.
edu/maps.php). Leuliette et al.'s analysis was based on mea-
surements from NASA's Ocean Topography Experiment
(TOPEX) and Jason satellite altimeters; results were cali-
brated using a model documented in Leuliette et al. (2004).
Satellite measurements can be obtained from NASA's online
database (NASA, 2006) (http://topex-www.jpl.nasa.gov/
science/data.html).
References
Burkett, V.R., D.B. Zilkoski, and D.A. Hart. 2005. Sea-
level rise and subsidence: Implications for flooding in New
Orleans, Louisiana. In: Subsidence observations based on
traditional geodetic techniques, and numerical models. U.S.
Geological Survey, National Wetlands Research Center.
Condrey, R., P. Kemp, J. Visser, J. Gosselink, D. Lindstedt,
E. Melancon, G. Peterson, and B. Thompson. 1995. Status,
trends, and probable causes of change in living resources in
the Barataria and Terrebonne estuarine systems. Thibodaux,
LA: Barataria-Terrebonne National Estuary Program.
Erwin, R.M. 2005. Atlantic sea level rise, lagoonal marsh
loss, and wildlife habitat implications. U.S. Geological
Survey. Accessed December 29, 2005.
Gill, S.K., andJ.R. Schultz. 2001. Tidal datums and
their applications. NOAA Special Publication NOS
CO-OPS 1.
Hartig, E.K., F. Mushacke, D. Fallen, and A. Kolker.
2000. A wetlands climate change impact assessment for the
metropolitan East Coast region. Draft for public review.
Leuliette, E.W., R.S. Nerem, G.T Mitchum, and D.P.
Chambers. 2006. Sea level change: 2006 release #3.
Accessed October 2006.
112
-------
Leuliette, E.W., R.S. Nerem, and G.T. Mitchum. 2004.
Calibration of TOPEX/Poseidon and Jason altimeter data
to construct a continuous record of mean sea level change.
Mar. Geod. 27(l-2):79-94.
Miller, L., and B.C. Douglas. 2004. Mass and volume
contributions to twentieth-century global sea level rise.
Nature 428:406-409.
NASA (National Aeronautics and Space Administration).
2006. Ocean surface topography from space. Updated
January 2006.
NOAA (National Oceanic and Atmospheric Administra-
tion). 2006. Sea levels online. Accessed October 6, 2006.
(home page); (data table)
NOAA. 2005. Population trends along the coastal United
States: 1980-2008.
NOAA. 2001. Sea level variations of the United States
1854-1999. NOAA Technical Report NOS CO-OPS 36.
Olff, H., J. De Leeuw, J.P. Bakker, R.J. Platermk, HJ. Van
Wijnen, and W. De Munck. 1997. Vegetation succession
and herbivory in a salt marsh: Changes induced by sea level
rise and silt deposition along an elevational gradient. J.
Ecol. 85:799-814.
Smith, R.A. 1980. Golden Gate tidal measurements. J.
Waterw. Port C. Div. 106(WW3):407-410.
Thieler, E.R., andE.S. Hammar-Klose. 1999. National
assessment of coastal vulnerability to sea-level rise: prelimi-
nary results for the U.S. Atlantic coast. U.S. Geological
Survey Open-File Report 99-593.
Titus, J., and C. Richman. 2001. Maps of lands vulnerable
to sea level rise: modeled elevations along the U.S. Atlantic
and Gulf coasts. Climate Res. 18:205-228.
U.S. Department of Transportation. 2003. Does sea level rise
matter to transportation along the Atlantic coast? In: U.S.
Department of Transportation, Center for Climate Change
and Environmental Forecasting. The potential impacts
of climate change on transportation.
USGS (United States Geological Survey). 1998. The
Chesapeake Bay: Geologic product of rising sea level. Fact
Sheet 102-98.
Williams, K., K.C. Ewel, R.P. Stumpf, F.E. Putz, and T.W.
Workman. 1999. Sea-level rise and coastal forest retreat on
the west coast of Florida, USA. Ecology 80(6):2045-2063.
113
------- |