oEPA
EPA 841 -R-16-113 1 December 2016
United States
Environmental Protection
Agency
National Lakes
Assessment 2012
A Collaborative Survey of
Lakes in the United States
-------
Acknowledgements
The U.S. Environmental Protection Agency (EPA) Office of Water (OW) would like to thank the many people who
contributed to this project. Without the collaborative efforts and support by state and tribal environmental agencies,
federal agencies, and other organizations, this assessment of lakes would not have been possible.
EPA would like to thank the steering committee, state and tribal environmental agencies, field crews, biologists,
taxonomists, laboratory staff, data analysts, program administrators, EPA regional coordinators, statisticians, quality
control staff, data management staff and many reviewers for their dedication and hard work. Your collective efforts
made this report possible. To the many hundreds of participants, EPA expresses its gratitude.
State, Tribal, Territory and Interstate Partners
Alaska Department of Environmental Conservation
Arizona Department of Environmental Quality
Blackfeet Tribe
California Department of Fish and Wildlife
Cheyenne River Sioux Tribe
Colorado Department of Public Health and
Environment
Connecticut Department of Energy and
Environmental Protection
Crow Nation
Delaware Department of Natural Resources and
Environmental Control
Fond du Lac Band of Chippewa Indians
Georgia Department of Natural Resources
Hawaii Department of Health
Hawaii Department of Land and Natural Resources
Idaho Department of Environmental Quality
Illinois Environmental Protection Agency
Indiana Department of Environmental
Management
Federal Partners
U.S. Department of Agriculture Forest Service
U.S. Department of Interior Geological Survey
Iowa Department of Natural Resources
Kansas Department of Health and Environment
Lac du Flambeau Band of Chippewa Indians
Leech Lake Band of Ojibwa
Maine Department of Environmental Protection
Maryland Department of Natural Resources
Menominee Indian Tribe of Wisconsin
Michigan Department of Environmental Quality
Minnesota Pollution Control Agency
Missouri Department of Conservation
Montana Department of Environmental Quality
Nevada Division of Environmental Protection
New Hampshire Department of Environmental
Services
New Jersey Department of Environmental
Protection
North Carolina Department of Environmental
Quality
North Dakota Department of Health
Ohio Environmental Protection Agency
Other Partners and Contractors
Oklahoma Water Resources Board
Oregon Department of Environmental Quality
Pennsylvania Department of Environmental
Protection
Rhode Island Department of Environmental
Management
Sisseton Wahpeton Oyate SiouxTribe
South Dakota Department of Environment &
Natural Resources
Spirit Lake Nation
Susquehanna River Basin Commission
Texas Commission on Environmental Quality
Turtle Mountain Band of Chippewa Indians
Utah Department of Environmental Quality
Vermont Department of Environmental
Conservation
Virginia Department of Environmental Quality
Washington State Department of Ecology
West Virginia Department of Environmental
Protection
Wisconsin Department of Natural Resources
BSA Environmental Services
Dynamac
EcoAnalysts, Inc.
Great Lakes Environmental Center, Inc.
Indiana University
Kansas Biological Survey
Michigan State University
Midwest Biodiversity Institute
Moss Landing Marine Labs
Oregon State University
SRA International
TetraTech, Inc.
University of Missouri
Wisconsin State Lab of Hygiene
The following people played a pivotal role and lent their expertise to data oversight and analysis in this project: Karen
Blocksom, Phil Kaufmann,Tom Kincaid,Tony Olsen, Steve Paulsen, Dave Peck, John Stoddard, John Van Sickle, and
Marc Weber from EPA Office of Research and Development; Richard Mitchell and Amina Pollard from EPA Office of
Water; Alan Herlihy from Oregon State University.
The National Lakes Assessment 2012 was led by Amina Pollard with significant programmatic help from Susan
Holdsworth, Marsha Landis, Sarah Lehmann, Richard Mitchell and Mimi Soo-Hoo (ORISE Participant) from EPA Office of
Water; Steve Paulsen from EPA Office of Research and Development; and EPA Regional Coordinators.
The NLA 2012 report was written by Amina Pollard with support from Sarah Lehmann. lEc/Crow Insight provided
tremendous graphics support for the report and the interactive data dashboards, technical editing, and layout
assistance. Comments from three peer reviewers improved this report.
The suggested citation for this document is: USEPA. 2016. National Lakes Assessment 2012: A Collaborative Survey of Lakes in the United States.
EPA 841-R-16-113. U.S. Environmental Protection Agency, Washington, DC. https://nationallakesassessment.epa.gov/
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Table of Contents
1 Executive Summary 1
2 Introduction 3
3 Design of the Lakes Survey 4
3.1 Choosing Indicators 4
3.2 Selecting Lakes 4
3.3 Field Sampling 6
3.4 Assessing 2012 Lake Conditions 7
3.4.1 Developing NLA-Derived Benchmarks 7
3.5 Estimating the 2012 Condition of the Population of Lakes 10
3.6 Differences Between the 2012 and 2007 Assessments 10
3.7 Assessing Changes in Condition Between 2007 and 2012 10
4 The Condition of Lakes in the United States 11
4.1 Chemical Condition Indicators 12
4.2 Physical Habitat Condition Indicators 15
4.3 Recreational Condition Indicators 18
4.4 Biological Condition Research Indicators 22
5 Comparing Conditions Across Ecoregions 24
6 Assessment Results and Future Data Uses 28
6.1 Current Condition and Change 28
6.2 Associations Between Stressors and Biological Condition 28
6.2.1 Relative Extent of Most Disturbed Condition 29
6.2.2 Relative Risk 29
6.2.3 Attributable Risk 29
6.3 Implications for Lake Managers 31
6.4 Next Steps for the National Surveys 31
7 Sources and References 33
Appendix A 34
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
0
-------
Tables and Figures
Figure 3.1 NLA 2012 Sampled Sites 5
Figure 3.2 NLA Field Crew Indicator Sampling Locations 6
Figure 3.3 Using Reference Lakes to Estimate Conditions 9
Figure 3.4 Distribution of Lakes among Size Classes in NLA 10
Figure 4.1 Interpreting Lake Condition Graphics 11
Figure 4.2 Trophic State National Condition Estimates 12
Figure 4.3 Phosphorus (Total) National Condition Estimates 13
Figure 4.4 Nitrogen (Total) National Condition Estimates 13
Figure 4.5 Oxygen (Dissolved) National Condition Estimates 14
Figure 4.6 Acidification National Condition Estimates 15
Figure 4.7 Riparian Vegetation Cover National Condition Estimates 15
Figure 4.8 Shallow Water Habitat National Condition Estimates 16
Figure 4.9 Lake Drawdown Exposure National Condition Estimates 17
Figure 4.10 Lakeshore Disturbance National Condition Estimates 17
Figure 4.11 Lake Habitat Complexity National Condition Estimates 18
Figure 4.12 Chlorophyll-a (Risk) National Condition Estimates 19
Figure 4.13 Cyanobacteria (Risk) National Condition Estimates 19
Figure 4.14 Microcystin (Risk) National Condition Estimates 20
Figure 4.15 Atrazine (Exceeds 4ppb) National Condition Estimates 20
Figure 4.16 Mercury (Total) National Condition Estimates 21
Figure 4.17 Mercury (Methyl) National Condition Estimates 21
Figure 4.18 Benthic Macroinvertebrates National Condition Estimates 22
Figure 4.19 Zooplankton National Condition Estimates 23
Figure 5.1 Ecoregion Conditions at a Glance 25
Figure 5.2 Phosphorus (Total) Ecoregion Condition Estimates 26
Figure 5.3 Lakeshore Disturbance Ecoregion Condition Estimates 27
Figure 6.1 Estimated Risk to Biota Caused by Stressors 30
(Fy
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States Vw'
-------
Oakes and reservoirs provide many environmental, economic, and public health benefits. We use lakes for
drinking water, energy production, food, and recreation, while fish, birds, and other wildlife rely on them for
habitat and survival. In the National Lakes Assessment (NLA), the U.S. Environmental Protection Agency (EPA)
and its partners surveyed a wide array of lakes representative of those found in the U.S., from small ponds and prairie
potholes to iarge iakes and reservoirs.
The National Lakes Assessment 2012: A Collaborative Survey of the Lakes in the United States presents the results of a
second evaluation of the biological, chemical, physical, and recreational condition of lakes in the United States,
the first having been conducted in 2007, During spring and summer of 2012,89 field crews sampled 1,038
lakes across the country. Each field crew used consistent procedures to sample benthic
macroinvertebrates (e.g., insect larvae, snails, and clams), zooplankton (small animals
in the water column), algal toxins, atrazine, and nutrients and to observe near-shore
habitat so that results could be compared across the country.These measured
values were compared to NLA benchmarks, which are points of reference used to
determine the proportion of lakes that are relatively high quality (least disturbed),
medium quality (moderately disturbed), and degraded (most disturbed) in condition.
NLA 2012 Condition
® The NLA indicates that nutrient pollution is common in U.S. lakes; 40% of lakes
have excessive levels of total phosphorus and 35% have excessive levels of total
nitrogen. Nutrient pollution is the most widespread stressor among those
measured in the NLA and can contribute to algae blooms and affect public
health and recreational opportunities in lakes.
• An algal toxin, microcystin, is detected in 39% of lakes, but concentrations
rarely reach moderate or high levels of concerns established by the World
Health Organization (<1% of lakes).
® The herbicide atrazine is detected in 30% of lakes, but concentrations
rarely reach the EPA level of concern for plants in freshwater (<1% of
lakes).
® We find that 31% of lakes have degraded benthic
macroinvertebrate communities, while 21 % of lakes have
degraded zooplankton communities. NLA exploratory analyses
indicate an association between nutrients and biological
condition, with lakes with phosphorus pollution likely also to
have a degraded biological condition.
NLA Change:
• A comparison of the 2007 and 2012 National Lakes
Assessments indicates little change between surveys, in
most cases, the percentage of lakes in degraded biological,
chemical and physical condition did not change at the
national scale over this five-year period.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
• One notable exception to this pattern was observed with algal toxin measures.
An analysis of cyanobacteria cell density, a measure of the density of cells that
could produce cyanotoxins, shows a statistically significant increase (+8.3%) in
the percentage of lakes in the most disturbed category between 2007 and 2012.
The NLA identified a significant increase in the detection of microcystin among
lakes in 2012 (+9.5%). However, concentrations of this algal toxin remained low
and rarely exceeded WHO recreational levels of concern (< 1 % of the population)
in both assessments.
• Another difference emerged through additional in-depth analyses of nutrient
data. While we did not observe changes in the condition categories, analysts
found a dramatic 18.2% decline in the percentage of oligotrophic lakes (<10
pg/L of total phosphorus) and an overall increase in the median concentration
of phosphorus across all lakes.
The NLA offers a unique opportunity to frame discussions and plan strategies for
the protection and restoration of lakes across the United States. Results of the
NLA provide a broad range of information that can help us better understand the
condition of lakes in the United States, some of the stressors affecting them, and
how stressors relate to local conditions. Whiie we explore associations between
these indicators, the NLA analysis presented in this report does not seek to explain
or identify the causes of degraded conditions or sources of stressors.
Additional information from the NLA, including assessment of conditions
at regional scales, differences between natural lakes and reservoirs, and an
opportunity to explore population-level results in an interactive dashboard, is
available online: https://www.epa.gov/national-aquatic-resource-surveys/nla.
% Rvi w , imi &
i7,, ytj &
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
O
-------
Introduction
©ealthy lakes enhance our quality of life. They support complex and important food web interactions and
provide habitat for many types offish and wildlife. Lakes contribute to a healthy economy: they are an
important draw for tourism and provide recreational opportunities for millions of Americans.
The National Lakes Assessment (NLA) is one of four statistical surveys that make up the National Aquatic Resource
Surveys (NARS).The NARS are implemented by the U.S. Environmental Protection Agency (EPA), states, and tribes to
provide nationally consistent assessments of surface waters in the U.S. For more information on NARS, visit https://
www.epa.gov/nationai-aquatic-resource-surveys.
The NLA is designed to answer the following questions about lakes across the United States.
1. What is the current biological, chemical, physical, and recreational condition of lakes?
a. What is the extent of degradation among lakes?
b. Is degradation widespread (e.g., national) or localized (e.g., regional)?
2. Is the proportion of lakes in the most disturbed condition getting better, worse, or staying the same over time?
3. Which environmental stressors are most strongly associated with degraded biological condition in lakes?
This brief report presents information from the second National Lakes Assessment (NLA 2012). It provides national-
scale assessments and also compares the condition of lakes to those from the earlier NLA 2007 conducted by EPA and
its partners. You can find results for regional scales and comparisons between natural lakes and reservoirs using our
interactive dashboard at https://nationallakesassessment.epa.gov/.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Design of the Lakes Survey
o
akes in the U.S. are as varied and unique as the landscapes surrounding them. The NLA includes examples
of all lake types, including lakes, ponds, and reservoirs on private, state, tribal, and federal lands across the
conterminous U.S.
3.1 CHOOSING INDICATORS
Lakes vary greatly and indicators selected to characterize them should represent their varied aspects. For the NLA
2012, several indicators were chosen to assess the chemical, physical, recreational, and biological condition of lakes.
Although there are many more factors that affect lakes or could be used to describe their condition, we believe these
indicators to be the most representative at a national scale (USEPA 2009).
The chemical characteristics of lake condition, such as nutrient levels and dissolved oxygen, create environments
essential for aquatic organisms to survive and grow. Chemical conditions in lakes affect the health of primary
producers (algae), zooplankton, macroinvertebrates, and fish. Chlorophyll-a was used as an indicator of trophic state
(productivity).
To address recreational and human health-related considerations, the NLA examined concentrations of the algal toxin
microcystin, along with cyanobacteria cell counts and chlorophyll-a concentrations as indicators of the potential for the
presence of algal toxins. Mercury in sediment was assessed because it bioaccumulates in the food chain. This survey also
measured atrazine in water samples. Atrazine is one of the most commonly used herbicides in the United States.
Physical indicators of lake condition evaluated for the NLA 2012 include conditions on the water's edge (lakeshore)
and in shallow water, measures of human disturbance, and drawdown (the natural or intentional lowering of lake
water levels). Healthy physical habitat affects biological communities in many ways, such as providing food and shelter
for aquatic wildlife and moderating the magnitude, timing, and pathways of water, sediment, and nutrient inputs into
lakes.
To evaluate the biological condition of lakes for the 2012 assessment, NLA analysts developed two new research
indicators, one based on benthic macroinvertebrate communities (bottom-dwelling animals without backbones) and
one using zooplankton (microsopic animals in the water column).
3.2 SELECTING LAKES
EPA used a statistical sampling approach incorporating survey design techniques to select lakes for this assessment.
This approach has been used in social science and health fields to determine the status of populations using a
representative sample of relatively few individuals.
The 1,038 lakes sampled were identified using a stratified random sampling technique called probability-based
sample design. In such a design, every lake in the target population has a known probability of being selected for
sampling. Site selection was controlled for lake size and spatial distribution to make sure that sample sites were
representative of lakes in the U.S., reflecting the full range in character and variation among lakes across the U.S.
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
O
-------
The USGS/EPA National Hydrography Dataset plus (NHDPIus), version 2 (http://www.horizon-systems.com/nhdplus/
NHDp!usV2_home.php) was used to compute the number of lakes throughout the U.S. The NHDPIus is a multi-layered
series of digital maps that show topography, area, flow, location, and other attributes of U.S. surface waters. The
NHDPIus V2 has 389,005 features listed that could potentially be lakes in the conterminous U.S.
To be included in the survey, a water body had to be either a natural or man-made freshwater lake, pond, or reservoir
greater than 2.47 acres (1 hectare), at least 3.3 feet (1 meter) deep, with a minimum quarter acre (0.1 hectare) of
open water. In addition, lakes were required to have a lakewater minimum residence time of one week (these latter
three criteria could only be determined at the site).The Great Lakes and the Great Salt Lake were not included in the
survey, nor were commercial treatment and/or disposal ponds, brackish lakes, or ephemeral lakes. After applying
these criteria, analysts estimate that 159,652 water bodies are considered lakes by the NLA 2012 definition and thus
comprise the target population.
Another determinant of lake inclusion was accessibility. In some cases, crews were either denied permission by the
landowner or unable to reach the lake because of safety concerns, such as sharp cliffs or unstable ridges. Using both
data from the crews'experience and pre-sampling reconnaissance, an estimated 30% or 47,833 lakes were found to be
inaccessible. This leaves 111,818 lakes that the NLA 2012 was able to assess and is also known as the larger inference
population.
The NLA 2012 collected data from 1,038 lakes selected from a stratified random sample based on ecoregion, state,
and surface area in the larger inference population (Figure 3.1). Consequently, throughout this report, percentages
reported for a given indicator are relative to the 111,818 lakes in the inference population. For example, if the
Figure 3.1: NLA 2012 Sampled Sites
I it' rfrexes i
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
condition is described as most disturbed for 10% of lakes nationally, this means that the number of lakes estimated
to be degraded for that indicator is 11,181 lakes. Findings from the 2007 and 2012 assessments indicate that the lakes
distributed across the conterminous U.S. are roughly equally split between man-made and natural in origin.
NLA site selection ensures that EPA can make unbiased estimates concerning the health of lakes and that it can
quantify the uncertainty of its estimates. The greater the number of sites sampled, the more confidence in the results.
The NLA population results are determined by first calculating the number of lakes in each condition class for each
indicator. Then the site weights from the probability design, which reflect the number of lakes each site represents
across the total population of NLA lakes, are summarized within condition classes to estimate the lakes in least,
moderate, and most disturbed condition. The number of sites included in the NLA 2012 allows EPA to determine the
percentage of lakes in the conterminous U.S. that exceed a benchmark of concern with 95% confidence.
3.3 FIELD SAMPLING
Throughout the summer of 2012, 89 field crews sampled lakes across the U.S. Sampling each iake took a full day.
To ensure consistency in collection procedures and to assure the quality of resulting data, the crews participated in
training, used standardized field methods and data forms, and followed strict quality control protocols (USEPA 2012).
At each lake site, crews collected samples at a standard single station located at the deepest point in the lake (or at
50m in deep lakes) and recorded shoreland and shoreline observations at ten stations evenly distributed around
the lake perimeter (Figure 3.2). At the mid-lake station (the deepest point of the lake), crews took depth profiles for
temperature, pH, and dissolved oxygen with a water probe. A Secchi disk was used to measure water clarity and the
depth at which light penetrates the lake (the euphotic zone). Crews collected vertically integrated water samples
Figure 3.2: NLA Field Crew Indicator Sampling Locations
Riparian Zone (l5m)
One indicator was sampled at the edge
of the littoral zone at IO locations:
Biological Benthic Macroinvertebrates
Nine indicators were sampled
at the deepest point:
Biological
Chemical
m ¦
"ffc -
0'"'r/'TJ titioral Zone
(10m) |
The conditions of the riparian
and shoreline zones were
observed for 5 indicators at
10 locations around the lake:
Physical Lake Drawdown Exposure
Lake Habitat Complexity
Lakeshore Disturbance
Riparian Vegetation Cover
Shallow Water Habitat
Zooplankton
Acidification
Nitrogen (Total)
Oxygen (Dissolved)
Phosphorus (Total)
Atrazine
Chlorophyll a (Risk)
Cyanobacteria (Risk)
Microcystin
X
%
Two indicators were
sampled from sediment
at the deepest point:
Human Use Mercury (Methyl)
Mercury (Total)
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
0
-------
1
s
II
Many of the measures
included in NLA are
natural components
of lakes. For example,
some level of a nutrient
like phosphorus is
necessary to support
lake communities,
while algal toxins like
microcystin occur
naturally in lakes. The
NLA explores whether
these measures are out
of balance compared
to expectations or
benchmarks.
from the euphotic zone to measure nutrients, chlorophyll-a, and the algal toxin
microcystin. Field crews used a fine mesh (50|jm) and coarse mesh (150|am)
plankton net to collect a vertically integrated zooplankton sample, and took a
sediment core to provide data on concentrations of mercury in sediments.
At the ten lake perimeter stations, crews collected data and information on the
physical characteristics that could affect habitat quality. Crews also collected
benthic macroinvertebrates from this area using a 500|jm D-frame net. Filtering
and other sample preparations took place on shore.
3.4 ASSESSING 2012 LAKE CONDITIONS
Many of the measures included in NLA are natural components of lakes. For
example, some level of a nutrient like phosphorus is necessary to support lake
communities, while algal toxins like microcystin occur naturally in lakes. The NLA
explores whether these measures are out of balance compared to expectations or
benchmarks.
NLA analysts reviewed the data for each indicator independently and split the
population into three categories - most disturbed (i.e., measures are out of
balance or degraded), moderately disturbed, and least disturbed (i.e., measures
are in balance or in good condition). These condition category terms were selected
for internal consistency and to more closely align with the relative nature of most
of the measures. More specifically, the observed value for an indicator at each
sampled lake was assessed against benchmark ievels to determine its condition.
The NLA 2012 uses two types of assessment benchmarks. The first type is the
fixed, literature-based benchmark based on values in the peer-reviewed scientific
literature. For example, a World Health Organization literature benchmark is
used to classify lakes into different algal toxin risk categories for recreation
in freshwaters.The second type is the NLA-derived benchmark based on the
distribution (i.e., the range of values) of an indicator derived from regional
reference lakes data (Figure 3.3). Appendix A provides general information about
each NLA 2012 indicator and the NLA 2012 Technical Report provides specific
details about benchmarks (USEPA 2016).
3.4.1 Developing NLA-Derived Benchmarks
Selecting Reference Lakes. In order to assess the condition of the country's
lakes, some measures are compared to benchmarks developed from the range of
values observed in a set of reference lakes. A reference lake in the NLA is a lake,
either natural or man-made, with attributes (such as water quality) that come as
close as practical to those expected in a natural state, i.e., a least disturbed lake.
Data from all sampled lakes are evaluated against reference screening criteria
to determine the final set of lakes used to characterize the reference condition
(steps 1 and 2 in Figure 3.3). Four groups of reference lakes are selected, one for
each biological condition (benthic macroinvertebrates and zooplankton), one for
nutrient condition, and one for physical habitat condition. Our expectations for
lake characteristics differ between temperate forests and xeric areas, so screening
criteria reflect the least disturbed conditions found in the different ecoregions.
Detailed information about reference selection, including ecoregions and
screening criteria, is in the NLA 2012 Technical Report (USEPA 2016). In refining
benchmarks for the NLA 2012, some 2007 benchmark values were revised;
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
O
-------
therefore, direct comparisons should not be made between 2012 results and those reported in 2007. For purposes of
identifying change in this report, 2007 results were recalculated based on new 2012 benchmarks.
When considering reference condition, it is important to remember that many areas in the U.S. have been altered, with
natural landscapes transformed by urban and suburban development, agricultural activities, and resource extraction.
To reflect the variability across the American landscape, least disturbed lakes diverge from the natural state by varying
degrees. For example, remote lakes like those in the upper elevation wilderness areas of Montana may be virtually
pristine, while the highest quality least disturbed lakes in other parts of the country, especially in urban or agricultural
areas, may exhibit various levels of human disturbance. For this reason, reference conditions might differ among
regions.The resulting reference lakes represent the analysis team's best effort at selecting lakes that are the least
disturbed in specific regions across the country.
Benchmarks. After lakes are screened and reference lakes are selected, NLA-derived
benchmarks are set for each of the regions against which the greater pool of target lakes
are compared (step 3 in Figure 3.3). For NLA, the results for each indicator are classified
as least disturbed, moderately disturbed, or most disturbed relative to the reference
conditions established for each ecoregion using the reference lakes.That is,"least
disturbed"denotes an indicator value similar to that found in reference lakes
(i.e., high quality or good condition);"most disturbed"denotes conditions
worse than most reference conditions (i.e., low quality, degraded, or poor
condition); and "moderately disturbed" indicates conditions that are in-
between these two states (i.e., medium quality or fair condition) (step
4 in Figure 3.3).
NLA-derived benchmarks were chosen from the range of values
(i.e., the distribution) of all the reference sites in a region for a
given indicator. Following established statistical approaches,
the NLA uses percentiles of the reference distribution to
determine benchmarks. Sites rate least disturbed when
indicator scores are as good as the best 75% of the
reference distribution. Sites rate most disturbed when
they score worse than the worst 5% of the reference
distribution. Moderately disturbed sites fall in
between.
3.5 ESTIMATING THE
2012 CONDITION OF THE
POPULATION OF LAKES
After indicators were characterized as
least disturbed, moderately disturbed, and
most disturbed, we used the sampled lake
information to develop inferences for the
population of lakes. During this step, the site
weights from the probability design, which
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
The NLA least disturbed, moderately disturbed,
and most disturbed designations are relative
to NLA 2012 benchmarks, not individual state
water quality standards, and do not replace the
assessment by states and tribes of the quality
of lakes relative to their specific water quality
standards under the Clean Water Act.
-------
Figure 3.3: Using Reference Lakes to Estimate Conditions
EPA designed its sample to be representative of the unique characteristics of lakes within nine differ-
ent ecoregions. For many of the biological, chemical, and physical NLA indicators, this representative-
ness is essential for an accurate understanding of our lakes. The steps below describe EPA's process
for determining the relative condition of lakes, which takes into account the differing landscapes
within each ecoregion. For simplicity, this infographic identifies the steps involved in estimating phos-
phorus conditions for lakes in the Northern Appalachian ecoregion. The steps for other ecoregions
and indicators are similar.
2. Analyze Data to Identify
Reference Lakes. For each of the 120
lakes, EPA scientists compared selected data
against reference screening criteria. This
process yielded 45 lakes considered relative-
ly undisturbed. EPA screened and added to
the list 26 additional lakes from the 2007
NLA, for a total of 71 reference lakes in the
Northern Appalachian ecoregion.
3. Calculate
Condition
Benchmarks.
EPA then analyzed the
distribution of phos-
phorus values among
reference lakes, to set
the benchmarks for the
condition categories.
Each dot below indicates the observed phosphorus level at a reference lake. Many of the dots overlap because
they have a similar value. The Vermont lake indicated below by a short vertical black line was one of 71
reference lakes in the Northern Appalachian ecoregion. It had a phosphorus level of 31 micrograms/liter (ug/L).
> ©
0
75% of reference lakes have phosphorus
levels below 14.5 ug/L in the Northern
Appalachian ecoregion and are considered to
be in least disturbed condition.
100
120
140
160
180 ug/L
" 5% of reference lakes have phosphorus levels above 22 ug/L
and are considered to be in the most disturbed category.
Those in between 14.5 ug/L and 22 ug/L (highlighted in
yellow) are considered moderately disturbed.
4. Assign Condition
Categories. Using
a regional benchmark,
EPA then determined the
phosphorus condition of
each randomly sampled
lake.
o o
oo
0 ug/L 20
40
60
80
100
120
140
160
180 ug/L
For instance, this New York lake was assigned to the "most disturbed" category because its phosphorus level of
55 ug/L was above the benchmark in that region. Of the 99 randomly sampled Northern Appalachian lakes, 27
met the criteria for the most disturbed condition for phosphorus. EPA also categorized 25 lakes as moderately
disturbed and 47 lakes as least disturbed for the same indicator.
5. Estimate the Condition of Lakes in the Ecoregion. Based on a weighted analysis of randomly
sampled lakes, EPA estimated the proportion of all Northern Appalachian lakes in each condition category. For instance,
EPA found that 31% of all Northern Appalachian lakes are designated as most disturbed for phosphorus. The
confidence interval for this estimate is ±20% (11% to 51%). EPA used similar procedures to assess the phosphorus
conditions in other ecoregions. Based on this analysis, EPA found that 40% of all lakes nationally are designated as most
disturbed for phosphorus. The confidence interval for this estimate is ±6% (34% to 47%). EPA is more certain about the
national estimate because it is based on information from a much larger number of lakes. Find all the national condition
estimates in Chapter 4.
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
Vermont
Pennsylvania o 0
1. Collect Data from a Sample of Lakes.
In 2012, NLA teams collected environmental samples
and observations from 120 lakes in the Northern
Appalachian ecoregion (see map). The
lakes comprised a large random sample
(99), supplemented by a smaller
set of hand-picked lakes
believed by EPA to be
least disturbed (21). Ohio
New Hampshire
- Massachusetts
Rhode Island
-------
reflect the number of lakes each site represents, are applied to estimate the percentage of lakes in each condition
category. This analysis provides a point estimate of condition and the 95th percentile confidence intervals around that
estimate. In the graphs throughout this report, the margin of error is depicted as narrow darkened lines on either side
of the bars and represents the confidence interval for the estimate. For national estimates, the margin of error around
the NLA findings is approximately ±5% (Brown et al. 2001). For more information on the probability design and how
sites are weighted, please see the NLA 2012 Technical Report (USEPA 2016).
3.6 DIFFERENCES BETWEEN THE 2012 AND 2007 ASSESSMENTS
We included several improvements to the National Lakes Assessment between 2007 and 2012. The design was
expanded to be more representative of lakes in the U.S. by including smaller lakes between 1 and 4 hectares of surface
area.This change in design increased the number of lakes assessed from approximately 50,000 in 2007 to 111,818 in
2012. We added atrazine as an indicator in NLA 2012. Additional modifications included using different zooplankton
sampling nets to improve our sampling method and collecting more sediment for mercury analyses.
3.7 ASSESSING CHANGES IN CONDITION BETWEEN 2007 AND 2012
Whenever possible, this report discusses changes in condition between 2007 and 2012. The smallest lakes in the 2012
assessment (1 -4 hectares) were not sampled in 2007. We adjusted for this by excluding the smaller size class from
the change analysis, so that the inference populations were equivalent between assessments (Figure 3.4). The 2007
and the 2012 data were assessed against the 2012 benchmarks so that the assessment endpoints were comparable.
Following these adjustments, NLA analysts compared the proportion of the inference population in each of the
disturbance categories to determine whether there was a change between the 2007 and the 2012 assessments.
Figure 3.4: Distribution of Lakes Among Size Classes in NLA
Lakes included in the
Lakes included in the analysis
87
142
173
225
411
73
162
211
684
NLA 2007
NLA 2012
1-4
Hectares
4-10
Hectares
10-20
Hectares
20-50
Hectares
>50
Hectares
Lakes included in the analysis LaKes mciuaea m
of 2012 condition change analysis
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
0.
-------
n this chapter, we describe the condition of lakes based on chemical, physical, human use, and two new
research biological indicators. Additionally, we compare results to NLA 2007 to indicate whether lake
conditions changed (questions 1a and 2).
Results for lake condition estimate the proportion of lakes in three condition classes (least disturbed, moderately
disturbed, and most disturbed) relative to either literature benchmarks or NLA-derived benchmarks.
Chapter 3 of this report and the NLA 2012Technical Report (USEPA 2016) explain the assumptions underlying the
analysis and give details on how these estimates were prepared. The estimated number of lake results are referred to
as population estimates and each estimate is accompanied by a confidence interval that conveys the level of certainty
in the estimate. Figure 4.1 describes how to interpret the lake condition graphics.
Figure 4.1: Interpreting Lake Condition Graphics
This figure describes how to interpret the graphics depicting the biological, chemical, physical, and recreational
condition in lakes as well as change since NLA 2007. The change analysis only applies to lakes 4 hectares and larger
because smaller lakes were not sampled in 2007.
Current Condition
The bars represent EPA's 2012 estimate for
the proportion of lakes in each condition
category - here, 15%.
Direction of Change
The slope graphs show the change from
2007 to 2012. Here, the gentle slope
indicates a change of 8 percentage points.
The sloped line will be used to display
trends in future assessments.
Condition Category
2012 Percentage of Lakes
0%
20% 40% 60% 80% 1
Most Disturbed*
Moderately Disturbed
Least Disturbed* -
Not Assessed
Magnitude of Change
The diamond shows the change estimate
and the line conveys the range of
uncertainty. Here, the percentage of lakes in
the most disturbed category increased by 8
percentage points with a confidence interval
of 4 to 13.
2007-2012
Change in % Points
30%
-40% -20% 0% 20% 40%
Statistical Significance
Statistically significant change within a
condition category is indicated by an
asterisk (*). Here, EPA is 95% confident
that the proportion of lakes in the least
disturbed condition decreased from 2007
to 2.012.
Confidence Intervals
The darker line represents the
confidence intervals reflecting the
margin of error around the point
estimate. In this case, EPA is 95% certain
that, in 2012, between 11% and 19% of
all lakes in the target population are in
the most disturbed category.
Good or Bad?
Falling to the left or right of the zero line
means something different for each
condition category. Here, the decrease in
percentage points for lakes designated as
least disturbed is undesirable, as is the
increase in lakes designated as most
disturbed.
Slope Graphs
Many lines will appear nearly flat
signaling that there is little change. The
light gray line shows the position of 50%.
-0
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
4.1 CHEMICAL CONDITION INDICATORS
The NLA 2012 chemical condition assessment is based on information about nutrient concentrations, oxygen content,
acidification, and trophic state (i.e., productivity) of lakes. In-lake measurements are compared either to reference
conditions developed from a set of reference lakes in each ecoregion or to nationally consistent benchmarks (oxygen,
acidification, and trophic state).
Trophic State. Trophic state is a common approach for classifying the biological productivity in lakes. Lakes with
high nutrient levels, high plant production rates, and an abundance of plant life are termed eutrophic, whereas lakes
that have low concentrations of nutrients, low rates of productivity, and generally low biological biomass are termed
oligotrophic. Lakes that fall in between these two states are called mesotrophic. Lakes naturally exist across all trophic
categories; however, hypereutrophic conditions are usually the result of human activity and can be an indicator of
stress conditions.
Eutrophication is a slow, natural part of lake aging, but today human influences can increase the amount of nutrients
entering lakes. Human activities such as poorly managed agriculture or suburbanization of watersheds can result in
high levels of nutrients reaching lakes. This can lead to accelerated eutrophication and related undesirable effects,
including nuisance algae, excessive plant growth, murky water, lower levels of dissolved oxygen (DO), odor, and fish
kills.
We use chlorophyll-a concentration as a surrogate for measuring algal biomass and thus to estimate the trophic
status of lakes. Based on published chlorophyll-a benchmarks, 10% of U.S. lakes are classified as oligotrophic, 35% are
mesotrophic, 34% are eutrophic, and 21% are hypereutrophic (i.e., most disturbed, Figure 4.2). An analysis of trophic
state shows no statistically significant difference in the percentage of lakes in the most disturbed (i.e., hypereutrophic)
category between 2007 and 2012.
Figure 4.2: Trophic State | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I I I I I I I I I I I
Most Disturbed
H— ?1%
-
Trophic State Eutrophic
Trophic State Mesotrophic
1
34%
^ 35%
Trophic State Oligotrophic
13- 10%
<
~
Not Assessed
0%
'
L_
Phosphorus and Nitrogen. For this assessment, nutrients in the form of total phosphorus and total nitrogen were
evaluated as indicators of the chemical condition of lakes. Phosphorus and nitrogen are critical nutrients required for
all life. In appropriate quantities, these nutrients power the primary algal production necessary to support lake food
webs. Phosphorus and nitrogen are linked indicators that jointly influence both the concentrations of algae in a lake
and the clarity of water. The naturally occurring levels of these indicators vary regionally, as does their relationship
with turbidity and algal growth. For phosphorus and nitrogen, lakes were assessed relative to regionally specific NLA-
derived benchmarks.
In many lakes, phosphorus is considered the limiting nutrient, meaning that the available quantity of this nutrient
controls the pace at which algae are produced. This also means that modest increases in available phosphorus can
cause very rapid increases in algal growth. Results indicate that 45% of lakes are in least disturbed condition, 15%
are in a moderately disturbed condition, and 40% are in the most disturbed condition for phosphorus (Figure 4.3).
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
An analysis of total phosphorus shows no statistically significant difference in the percentage of lakes in the most
disturbed category between 2007 and 2012.
While there has been no detectable change in the proportion of lakes in each condition category, additional in-depth
analyses of the NLA data indicate striking differences in the distribution of phosphorus concentrations, particularly at
the low end of the phosphorus gradient (Stoddard et al. 2016). Increases in phosphorus in previously low phosphorus
lakes raised the median concentration across all lakes from 20 pg/L in 2007 to 37 pg/L in 2012. Equally striking is the
dramatic decline in the percentage of naturally oligotrophic lakes (<10 pg/L ofTP L-1), where the proportion changed
from 24.9% to 6.7% of the population. Please see Stoddard et al. (2016) for details of the analysis and discussion of
potential causes for these differences between assessments.
Figure 4.3: Phosphorus (Total) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I I I I I I I I I I I
Most Disturbed
! an%
Moderately Disturbed
15%
Least Disturbed
Not Assessed
No Observed Lakes
N/A
N,
A
Other lakes are limited by nitrogen. In these lakes, modest increases in available nitrogen might yield the same effects
that increases in phosphorus do elsewhere. NLA 2012 indicates that 41 % of lakes exhibit a least disturbed condition,
25% are in a moderately disturbed condition, and 35% are in the most disturbed condition for nitrogen (Figure 4.4). An
analysis of total nitrogen shows no statistically significant difference in the percentage of lakes in the most disturbed
category between 2007 and 2012.
Figure 4.4: Nitrogen (Total) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
iii i i i i i
Most Disturbed I 35%
Moderately Disturbed 25%
Least Disturbed 41%
Not Assessed
+-
H
N/A
m
Dissolved Oxygen. Dissolved oxygen, or DO, is considered an
important measurement of water quality because
it is essential for aquatic communities. Without
oxygen, a lake would be devoid offish and
macroinvertebrates. Aquatic organisms have
differing DO requirements for optimal growth
and reproduction. Changes in DO levels can
occur for a variety of reasons, including water
temperature, wind action, and the amount
of aigae and aquatic plants in the lake.
Eighty-eight
percent of lakes have high
levels of dissolved oxygen
(DO), which is essential for
healthy aquatic communities.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Eighty-eight percent of lakes have high levels of surface (epilimnetic) DO (are in the least disturbed condition); 8%
are moderately disturbed; 2% are in the most disturbed condition; and 2% of lakes were not assessed (Figure 4.5). An
analysis of DO shows no statistically significant difference in the percentage of lakes in the most disturbed category
between 2007 and 2012.
Figure 4.5: Oxygen (Dissolved) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
i L_ L
Most Disturbed
3" 2%
t
Moderately Disturbed
8%
Least Disturbed
—(— aa%
-
Not Assessed*
} 2%
~
* Reflects a statistically significant change between 2007 and 2012(95% confidence).
Acidification, Lake acidification can be an important indicator of lake condition. Acid rain and acid mine drainage
are major sources of acidifying compounds that can change the pH of lake water, impacting fish and other aquatic life.
Acid-neutralizing capacity (ANC) serves as an indicator of sensitivity to changes in pH.The ANC of a lake is determined
by the soil and underlying geology of the surrounding watershed. Lakes with high levels of dissolved bicarbonate
ions (e.g., limestone watersheds) are able to neutralize acid depositions and buffer the effects of acid rain. Conversely,
watersheds that are rich in granites and sandstones contain fewer acid-neutralizing ions and have iow ANC; these
systems have a predisposition to acidification. Maintaining stable and sufficient ANC is important for fish and aquatic
life because ANC protects or buffers against pH changes in the water body. Most aquatic organisms function at the
optimal pH range of 6.5 to 8.5. Sufficient ANC in surface waters will buffer acid rain and prevent pH levels from straying
outside this range. In naturally acidic lakes, the ANC may be quite low, but the presence of natural organic compounds
in the form of dissolved organic carbon, or DOC, can mitigate the effects of pH fluctuations.
Results indicate that almost all of the nation's lakes - 97% - can be classified in the least disturbed condition for
acidification (Figure 4.6). A 2012 vs. 2007 change analysis was not conducted for acidification.
National Lakes Assessment 2012 I A Collaborative
of Lakes in the United States
-------
Figure 4.6: Acidification ] National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I I I i I I I I I I I
Most Disturbed
0%
N/A
N
Moderately Disturbed
2%
N/A
N
A
Least Disturbed
97%
N/A
N
A
+
Natural Organic Acid
0%
N/A
N
A
Not Assessed
o
o
z
sserved Lakes
N/A
IN,
/a
4.2 PHYSICAL HABITAT CONDITION INDICATORS
The condition of lakeshore habitats provides important information relevant to lake biological health. For the NLA
2012, physical habitat condition was assessed based on observations of five indicators: riparian (lakeshore) vegetation
cover, littoral (shallow water) habitat, lake drawdown (lowering of lake levels), habitat disturbance (extent and
intensity of human activity), and habitat complexity (a combined index of condition at the land-water interface).
Although lake drawdown is a new indicator for 2012, data are available from 2007 that allowed analysts to develop
results from both periods.
Riparian Vegetation Cover. Evaluation of riparian or lakeshore vegetation cover is based on observations of three
layers of vegetation - understory grasses and forbs, mid-story non-woody and woody shrubs, and over-story trees.
Although generally shorelines are in better condition when vegetation cover is high in all layers, not all three layers
occur in all areas of the country. For example, in the Northern Plains there is typically no natural over-story tree cover;
in the West, steep rocky shores are the norm for high-mountain or canyon lakes.These natural features have been
factored into the calculation of the riparian vegetation cover indicator.
Nationally, 48% of lakeshore habitats are in a least disturbed condition; 23% are in a moderately disturbed condition;
28% are in most disturbed condition; and <1% of lakeshores were not assessed (Figure 4.7). An analysis of riparian
vegetation cover indicates no significant difference in the percentage of lakes in the most disturbed category between
2007 and 20121.
Figure 4.7: Riparian Vegetation Cover | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
Most Disturbed
!
4-
¦
Moderately Disturbed
Least Disturbed
Not Assessed*
23%
I - ™lH— JA%
0%
* Reflects a statistically significant change between 2007 and 2012(95% confidence).
' NLA 2007 did not require physical habitat observations on iakes >5,000 ha. NLA 2012 increased this limit to > 10,000 ha. This slight methodological change is associated
with a small, significant increase in the proportion of lakes assessed for habitat measures.
G)
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Shallow Water Habitat. The shallow water habitat indicator examines the quality of the shallow edge of the lake
by using data on the presence of living and non-living features such as overhanging vegetation, aquatic plants, large
woody snags, brush, boulders, and rock ledges. Lakes with greater and more varied shallow water habitat are typically
able to more effectively support aquatic life because they have many complex ecological niches. Like the riparian
or lakeshore habitat indicator, the shallow water indicator is related to conditions in reference lakes and is modified
regionally to account for differing expectations of natural condition.
The NLA 2012 finds that shallow water habitats are in a least disturbed condition in 55% of U.S. lakes; are moderately
disturbed in 27% of lakes; and are most disturbed in 18% of lakes (Figure 4.8). An analysis of shallow water habitat shows
no statistically significant difference in the percentage of lakes in the most disturbed category between 2007 and 2012.
Figure 4.8: Shallow Water Habitat | National Condition Estimates
Condition Category
2012 Percentage of Lakes
2007-2012 Change in % Points
0%
20% 40% 60% 80% 100%
i i i i i
-40% -20% 0% 20% 40%
i
Most Disturbed I 18%
Moderately Disturbed 27%
Least Disturbed
55%
Not Assessed* 1 0%
I
t
1
* Reflects a statistically significant change between 2007 and 2012 (95% confidence).
Lake Drawdown Exposure. Lake drawdown can occur in both natural lakes and reservoirs. It can be the result of
natural processes, such as periodic drought, or the result of direct manipulation of water levels for lake management
purposes. Changing or significantly lowered iake water levels can adversely affect physical habitat conditions in and
around the lake and therefore can also have an impact on biological communities.The NLA lake drawdown indicator
measures whether water levels are lower than their full-lake stage.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Most U.S. lakes (79%) showed least disturbed levels of lake drawdown in 2012; 14% showed moderately disturbed
levels; and 6% showed most disturbed levels of lake drawdown (Figure 4.9). An analysis of lake drawdown exposure
shows a statistically significant lower (-12.9%) percentage of lakes in the most disturbed category between 2007 and
2012.
Figure 4.9: Lake Drawdown Exposure | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I i I I I I I L I I i _
Most Disturbed*
3* 6%
Moderately Disturbed*
14%
~ *¦
Least Disturbed*
— 79%
-4-
H
Not Assessed*
0%
i
>
* Reflects a statistically significant change between 2007 and 2012 (95% confidence).
Lakeshore Disturbance. The lakeshore disturbance indicator reflects direct human alteration of the lakeshore
itself. These disturbances can range from minor changes, such as the removal of a few trees to develop a picnic
area, to major alterations, such as the construction of a large lakeshore residential complex. The effects of lakeshore
development on the quality of iakes include excess sedimentation, loss of native plants, alteration of native plant
communities, loss of vegetation structure and complexity, and modifications to substrate types. These impacts, in
turn, can negatively affect fish, wildlife, and other aquatic communities.
Across the lower 48 states, 29% of lakeshores are in the least disturbed condition; 53% show moderately disturbed
levels; 18% are most disturbed; and 1% of lakeshores were not assessed (Figure 4.10). An analysis of lakeshore
disturbance shows no statistical difference in the percentage of lakes in the most disturbed category between 2007
and 2012.
Figure 4.19: Lakeshore Disturbance | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I I I I I I I I I I, I
Most Disturbed
Moderately Disturbed
18%
]
53%
Least Disturbed
—~""J
Not Assessed
1 1%
|
~
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Lake Habitat Complexity. The habitat complexity indicator combines lakeshore and shallow water indictors
described above to estimate the amount and variety of all cover types at the water's edge (on land and in water). This
indicator is compared to NLA-derived regional benchmarks.
For lake habitat complexity at the iand-water interface, 43% of lakes are in a least disturbed condition; 28% are in
moderately disturbed condition; and 29% are in a most disturbed condition (Figure 4.11). An analysis of lake habitat
complexity shows no statistically significant difference in the percentage of lakes in the most disturbed category
between 2007 and 2012.There were fewer lakes in the least disturbed category and more lakes in the moderately
disturbed category.
Figure 4.11: Lake Habitat Complexity j National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I 1 I I
Most Disturbed
;
Moderately Disturbed*
28%
Least Disturbed*
—Ir— 43%
—
Not Assessed*
0%
~
* Reflects a statistically significant change between 2007 and 2012(95% confidence).
4.3 RECREATIONAL CONDITION INDICATORS
Lakes are used for a wide variety of recreational purposes, including swimming, waterskiing, windsurfing, fishing, and
boating. Contaminants in lakes can pose a potential threat to humans, pets, and wildlife.The NLA 2012 assessed algal
toxins, mercury in sediment, and atrazine as indicators of human use or recreational condition in U.S. lakes.
Algal Toxins. Algae and cyanobacteria are a natural part of freshwater ecosystems. Eutrophication in lakes often
results in conditions that favor their growth. Many algal blooms are unsightly, but not toxic. However, some blooms of
cyanobacteria can be harmful to people and animals. Exposure to cyanobacteria toxins may produce skin rashes, eye
irritations, respiratory symptoms, gastroenteritis, and liver and kidney failure.
The World Health Organization (WHO) established recreational exposure risk guidelines for chlorophyil-a,
cyanobacterial cell counts, and microcystin.These literature benchmarks were used in the NLA to determine risk of
exposure to algal toxins. It is important to note that chlorophyll-o concentrations and cyanobacteria cell counts serve
as proxies for the potential presence of algal toxins. A lake that is in least disturbed condition exhibits a low risk of
exposure; a lake in a most disturbed condition has a high exposure potential.
The algal toxin microcystin was detected in
39% of lakes, but very rarely at levels that
represent moderate or high levels of exposure
risk to the recreating public.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
WARNING
KEEP OUT
-------
Using the WHO literature benchmarks, the level of risk associated with exposure to algal toxins varies by indicator.
For chlorophyll-o, 15% of lakes are in the most disturbed condition (i.e., pose a high risk of exposure according to the
WHO literature benchmarks), 34% pose a moderate risk, and 51% are in the least disturbed condition (Figure 4.12).
An analysis of chlorophyll-a shows no statistical difference in the percentage of lakes in the most disturbed category
between 2007 and 2012.
Figure 4.12: Chlorophyll-# (Risk) j National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I i j i i i I .i i i i
Most Disturbed
Moderately Disturbed
Least Disturbed
Not Detected
Not Assessed
15%
•
-
mm
h
34%
-
-
I 51%
I 1%
I 1%
¦
~
Using cyanobacteria cell counts as an indicator of risk for exposure to algal toxins, 15% of lakes are in the most
disturbed condition (i.e., pose a high risk of exposure to the public); 23% indicate moderately disturbed condition;
61% are in the least disturbed condition; and 1% were not assessed (Figure 4.13). An analysis of cyanobacteria cell
density shows a statistically significant increase (+8.3%) in the percentage of lakes in the most disturbed category
between 2007 and 2012.
Figure 4.13: Cyanobacteria (Risk) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
Most Disturbed*
'
~
15%
Moderately Disturbed
23%
Least Disturbed*
H 61%
Not Assessed*
\ 1%
»
* Reflects a statistically significant change between 2007 and 2012(95% confidence).
National Lakes Assessment 2012 I A Collaborative
of Lakes in the United States
-------
Microcystin, a direct measure of an algal toxin, was detected in 39% of lakes. Less than one percent of lakes are in
the most and moderately disturbed condition (i.e., have a high or moderate risk of exposure) and 99% are either
least disturbed, with a low risk of exposure, or show no detection of microcystin (Figure 4.14). Although there was a
significant increase in the detection of microcystin (+9.5%), the analysis of microcystin shows no statistically significant
difference in the percentage of lakes in the most disturbed category between 2007 and 2012.
Figure 4.14: Microcystin (Risk) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% 40% -20% 0% 20% 40%
I I I J I I I I | I I
Most Disturbed
! o%
L
w
Moderately Disturbed
0%
Least Disturbed*
Not Detected *
Not Assessed
-
—f— 6(1%
} 1%
* Reflects a statistically significant change between 2007 and 2012(95% confidence).
Atrazine, Atrazine, one of the most widely used agricultural herbicides in the U.S., is applied before and after planting
to control broadieaf and grassy weeds. According to studies by the U.S. Geological Survey, atrazine is the most
frequently detected pesticide in streams and shallow groundwater (Gilliom et al. 2006), Atrazine can also end up in
lakes. We added atrazine to the NLA 2012 to examine the frequency of detection of atrazine in lakes across the U.S.
during the summer index period.
Atrazine was detected in 30% of lakes and not detected in 70% of lakes (least disturbed). Less than one percent of
lakes are in the most disturbed condition (where atrazine has a high risk of affecting plant communities) according to
the EPA-proposed level of concern for plants in freshwaters (4 ppb) (Figure 4.15). Because atrazine is a new indicator in
the NLA 2012, a comparison with 2007 is not possible for atrazine measures.
Figure 4.15: Atrazine (Exceed 4 ppb) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I I I I I I I I I I I
Most Disturbed
0%
Moderately Disturbed
30%
Least Disturbed
—(— 70%
Not Assessed
0%
Mercury in sediment Mercury is found in many rocks, including coal. When coal is burned, mercury is released into
the environment. Some of the mercury in the air eventually settles into water or is washed into water. Once mercury
is deposited, certain microorganisms can change it into methylmercury, a highly toxic form that builds up in fish,
shellfish, and animals (including humans) that eat fish. Mercury exposure at high levels can harm animal behavior,
reproduction, growth and development.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
20
-------
For total mercury in sediment, 21% of lakes are in the least disturbed condition; 51% indicate moderately disturbed
condition; 26% are in the most disturbed condition; and 2% of lakes were not assessed (Figure 4.16), A change analysis
was not possible for total mercury because of different sampling protocols between 2007 and 2012.
Figure 4.16: Mercury (Total) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
Most Disturbed
?b%
Moderately Disturbed
51%
Least Disturbed
Not Assessed
—i— ?i%
} 2%
n
For methylmercury in sediment, 31% of lakes are in the least disturbed condition; 28% are moderately disturbed;
40% are in the most disturbed condition; and 1 % of lakes were not assessed (Figure 4,17). A change analysis was not
possible for methylmercury because of different sampling protocols between 2007 and 2012. Additional research is
needed to evaluate the relationships between mercury in sediments and other lake conditions.
Figure 4.17: Mercury (Methyl) | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
I I I I I | I I I I 1
Most Disturbed
I —I— 40%
Moderately Disturbed
28%
Least Disturbed
31%
Not Assessed
$ 1%
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
4.4 BIOLOGICAL CONDITION RESEARCH INDICATORS
Biological indicators are commonly used in stream and river water quality assessment programs, but they are not
typically used in lake monitoring programs. Aquatic organisms integrate multiple water quality factors over time, such
that communities reflect their environment. Because of this assessment benefit, we developed two new biological
research indicators for NLA 2012. The sensitivity of these new indicators to environmental factors (e.g., the effect of
soft sediment versus cobble substrate) and biological variables (e.g., the effect offish predation) is an area of active
research. For this reason, we present them as research indicators in this report. In future assessments, we may revise
and refine these biological indicators to reflect new understanding.
The biology of a lake can be characterized by the presence, number, and diversity of macroinvertebrates, algae,
vascular plants, and other organisms that together provide information about the health and productivity of the
lake ecosystem.The new NLA 2012 biological condition assessment indicators are based on information from two
biological communities: benthic macroinvertebrates in the littoral (shallow water) zone, and zooplankton from a
pelagic (open water) zone. In order to assess biological health, NLA analysts combined several measures into indices.
Benthic Macroinvertebrates. Benthic macroinvertebrates include aquatic insects in their larval stage, small aquatic
mollusks (snails and clams), crustaceans (e.g., crayfish), aquatic worms, and leeches.They live on and under the rocks,
sediments, and vegetation at the bottom of lakes. These organisms were selected as indicators of biological condition
because they spend most of their lives in water and are thought to respond to human disturbance. Given their broad
geographic distribution, abundance, ease of collection, and connection to fish and other aquatic organisms (e.g., as a
source of food), these organisms may serve as good indicators of the biological quality of shoreline habitats in lakes.
To create the benthic invertebrate indicator, NLA analysts selected measures of six different aspects of
macroinvertebrate community structure: taxonomic composition; taxonomic diversity; feeding groups; habits/
habitats; taxonomic richness; and pollution tolerance. The measures chosen for each of these aspects vary among
ecoregions and are described in detail in the NLA 2012 Technical Report (USEPA 2016).
According to the benthic macroinvertebrate indicator, we estimate that 33% of U.S. lakes are in least disturbed
condition; 26% are in moderately disturbed condition; 31% are in a most disturbed condition; with 11% of lakes
not assessed (Figure 4.18). An analysis of benthic invertebrates shows no statistically significant difference in the
percentage of lakes in the most disturbed category between 2007 and 2012.
Figure 4.18: Benthic Macroinvertebrates | National Condition Estimates
Condition Category 2012 Percentage of Lakes 2007-2012 Change in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40%
__l | | | i |
31%
Most Disturbed
Moderately Disturbed 26%
Least Disturbed 33%
Not Assessed
11%
n
Zooplankton. Zooplankton are small animals in the water column that constitute an important element of the
aquatic food web. These organisms serve as an intermediary species, transferring energy from algae (primary
producers) to larger invertebrate predators and fish. Zooplankton are sensitive to changes in the lake ecosystem.
Given their broad geographic distribution, abundance, ease of collection, and connection to fish and other aquatic
organisms, these organisms may serve as good indicators of the biological quality of open water in lakes. To determine
the zooplankton indicator, NLA analysts selected six measures of community structure: abundance, taxonomic
richness, trophic guild, and three taxonomic measures (cladoceran, copepod, and rotifer). The specific metrics chosen
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
0.
-------
for each of these characteristics varied among the ecoregions and are described in detail in the NLA 2012 Technical
Report.
According to the zooplankton indicator, 53% of lakes in the U.S. are in a least disturbed condition; 27% are in a
moderately disturbed condition; and 21% are in a most disturbed condition (Figure 4.19). A change analysis was not
possible for the zooplankton metric because different sampling protocols were used in 2007 and 2012.
Figure 4.19: Zooplankton | National Condition Estimates
Condition Category
2012 Percentage of Lakes
2007-2012
Change in % Points
0% 20% 40% 60% 80% 100%
I I I I I I
-40% -20% 0% 20% 40%
I I I I I
Most Disturbed
Moderately Disturbed
Least Disturbed
Not Assessed
! 71%
27%
SS%
0%
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
5
Comparing Conditions Across Ecoregions
o
he design of the NLA allows us to explore indicators regionally as well as nationally (Question 1b). In this
chapter, we provide a simple example of how new NLA graphics can be used to consider questions about
lakes at the scale of ecological regions.
Results for all of the NLA 2012 indicators for each ecoregion are presented in a size plot (Figure 5.1), Similar to heat
maps that use color to represent data values, a size plot provides an immediate visual summary of complex data
sets. In this case, the size of each box represents the proportion of lakes in most disturbed condition for the related
indicator - the iarger the box, the more lakes NLA found to be in most disturbed condition. This graphic allows us to
look at patterns in data across ecoregions and indicators at a glance.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Figure 5.1: Ecoregion Conditions at a Glance
CPL: Coastal Plains NAP: Northern Appalachians NPL: Northern Plains
Comparing NLA results across ecoregions arid indicators can illuminate patterns that might otherwise remain hidden.
Below, read more about the value of ecoregion analysis and a graphical tool for high-level comparisons.
What Are Ecoregions?
Ecoregions are areas that contain similar environ-
mental characteristics and are defined by common
natural characteristics such as climate, vegetation,
soil type, and geology. Because of the diversity of
landscapes, it is important to assess waterbodies in
their own geographical setting. The NLA was de-
signed to report findings on an ecoregional scale.
By looking at lake conditions in ecoregions, decision-
makers can begin to understand patterns based
on landform and geography, and whether key lake
condition challenges are isolated or widespread.
The maps shown to the right represent the regions
displayed in the size plot below.
SAP: Southern Appalachians
SPL: Southern Plains
TPL:
Temperate Plains
Wt
UMW: Upper Midwest
WMT: Western Mountains
XER: Xeric
High-Level View of Conditions Across Ecoregions and Indicators
This graphic summarizes the percentage of lakes in most disturbed condition for each indicator and ecoregion. Use
it to spot high-level differences, similarities and outliers. Then dig deeper with the interactive graphics at the NLA
website: https://www.epa.gov/national-aquatic-resource-surveys/nla.
CPL
NAP
NPL
SAP
SPL
TPL
UMW
WMT
XER
Biological
Benthic Invertebrates
¦
¦
¦
¦
¦
¦
¦
¦
¦
Zooplankton
¦
¦
¦
¦
¦
¦
¦
¦
¦
Chemical
Acidification
Nitrogen (Total)
¦
¦
¦
¦
¦
¦
¦
¦
¦ -n
Oxygen (Dissolved)
¦
¦
¦
¦
¦
• ^
Phosphorus (Total)
¦
¦
¦
¦
¦
¦
¦
¦
¦
Trophic State
¦
¦
¦
¦
¦
¦
¦
¦
¦
Human Use
Atrazine (Detected)
¦
¦
¦
¦
¦
¦
¦
¦
¦
Atrazine (Exceeds 4ppb)
•
Chlorophyll A (Risk)
¦
¦
¦
¦
¦
¦
¦
Cyanobacteria (Risk)
¦
¦
¦
¦
¦
¦
¦
¦
Mercury (Methyl)
¦
¦
¦
¦
¦
¦
¦
¦
¦
Mercury (Total)
¦
¦
•
¦
¦
¦
¦
Microcystin (Detected)
¦
¦
¦
¦
¦
¦
¦
¦
¦
Microcystin (Risk)
¦
•
Physical
Lake Drawdown Exposure
¦
•
¦
¦
¦
¦
¦
¦
¦
Lake Habitat Complexity
¦
¦
¦
¦
¦
¦
¦
¦
¦
Lakeshore Disturbance
¦
¦
¦
¦
¦
¦
¦
¦
¦
Riparian Vegetation Cover
¦
¦
¦
¦
¦
¦
¦
¦
¦
Shallow Water Habitat
¦
¦
¦
¦
¦
¦
¦
¦
¦
The area of the squares »»%
~ >0% ¦—
is proportionate to the s%
point estimate for lakes
in the most disturbed condition.
Large proportions of lakes in each ecore-
gion are in the most disturbed condition
for Nitrogen (Total), in contrast with
consistently low proportions for Oxygen
(Dissolved). We see a similar pattern
when comparing the two microcystin
indicators.
The column for Northern Plains (NPL)
shows a high proportion of lakes in the
most disturbed condition for many physi-
cal indicators, a pattern that stands out
from other ecoregions. What patterns do
you see?
-01
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Next, we can explore regional differences in more detail. Nutrient pollution is common across the United States, but
levels vary widely among ecoregions. For total phosphorus (Figure 5,2), NLA results indicate that the Northern Plains
has the highest proportion of lakes in the most disturbed condition (80%). The Northern Plains and the Southern
Appalachians have statistically significant higher proportions of lakes in a most disturbed condition for phosphorus
than is observed across the national data set (the national confidence interval, surrounding the national point
estimate of 40%, is noted by the solid gray band that falls between 34% and 47% in Figure 5.2). The Upper Midwest
and the Western Mountains have statistically significant lower proportions of lakes in the most disturbed category for
phosphorus than observed nationally, A statistically significant increase (+27%) in the proportion of lakes in the most
disturbed condition is observed in the Northern Appalachians.
Figure 5.2: Phosphorus (Total) | Ecoregion Condition Estimates
Showing Data by Subpopulation -»• 2012 % of Lakes (Most Disturbed) 2007-12 Change in % Points
-40% -20% 0% 20% 40% 60% 80%
Coastal Plains
Northern Appalachians*
Northern Plains
Southern Appalachians
Southern Plains
Temperate Plains
Upper Midwest
Western Mountains
Xeric
0% 20% 40% 60% 80% 100%
50%
80%
* Reflects a statistically significant change between 2007 and 2012(95% confidence).
' i
Some NLA results show big
differences among regions in
the U.S. For example; the
proportion of lakes in the
most disturbed condition
for total phosphorous
ranges from 80% in the
Northern Plains to 17% in
the Western Mountains.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Levels of lakeshore disturbance also vary among ecoregions (Figure 5.3).The Northern Plains has the highest
proportion of lakeshores in the most disturbed condition (69%), and this is a statistically significant difference from
national levels (the national confidence interval, which surrounds the national point estimate of 18%, is noted by
the solid gray band that falls between 13% and 22% in Figure 5.3). The Northern Appalachians (6%) and the Western
Mountains (6%) had statistically significant lower proportions of lakeshores in the most disturbed condition category
compared to national levels. A significant decrease (-9%) in the proportion of lakeshores in the most disturbed
condition is observed in the Western Mountains.
Figure 5.3: Lakeshore Disturbance | Ecoregion Condition Estimates
Showing Data by Subpopulation ~ 2012 % of Lakes (Most Disturbed) 2007-12 Charge in % Points
0% 20% 40% 60% 80% 100% -40% -20% 0% 20% 40% 60% 80%
I I ! I I i I I I I I I I
Coastal Plains
•
~
Northern Appalachians
3-
6%
Northern Plains
~
Southern Appalachians
11%
a
Southern Plains
4
10 /o
'—
~
Temperate Plains
4
?
Upper Midwest
1
2%
4-
Western Mountains*
3-
6%
Xeric
* Reflects a statistically significant change between 2007 and 2012(95% confidence).
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
-------
Ohe NLA offers a unique opportunity to frame discussions and planning strategies across jurisdictional lines.
It is an evaluation of the collective successes of management efforts to protect, preserve, or restore water
quality. The NLA condition results provide information to understand the condition of lakes in the U.S.
6.1 CURRENT CONDITION AND CHANGE
While many lakes are in good condition, the NLA indicates that a substantial portion of lakes in the U.S. are in a most
disturbed condition for nutrients; 40% of lakes contain excessive total phosphorus concentrations and 35% of lakes
have excessive nitrogen concentrations. NLA indicates that 21% of the population of lakes are hypereutrophic. For
research biological measures, we find that 31% and 21% of lakes are in the most disturbed condition based on benthic
macroinvertebrate and zooplankton, respectively.
An analysis of change between 2007 and 2012 assessments indicates that the proportion of lakes in the most
disturbed condition did not change significantly for most indicators on a national scale. The lack of change might
be because lakes are slow to change. There were three exceptions to this pattern. First, there was a significant
8.3% increase in the percentage of lakes in the most disturbed condition based on cyanobacteria cell density, an
indicator of risk to exposure to algal toxins. There was also a significant 9.5% increase in the detection of an algal
toxin, microcystin, among lakes in 2012; however, concentrations of this algal toxin remained low and exceeded
WHO recreational levels of concern in both assessments in less than 1% of lakes. The third exception was a significant
12.9% decrease in the percentage of lakes in the most disturbed category for lake drawdown exposure. We encourage
additional investigation into the causes and consequences of these noted changes.
6.2 ASSOCIATIONS BETWEEN STRESSORS AND BIOLOGICAL CONDITION
In this section, we evaluate which most disturbed environmental conditions, or stressors evaluated in NLA, are
associated with degraded biological condition in lakes (Question 3).This simple analysis relies on the new research
indicators for biological condition and could be updated as we refine our understanding of the biological measures.
We rank these stressors in terms of the benefits expected to be derived from reducing or eliminating them.
For the NLA, analysts applied three approaches to rank stressors. The first looks at the extent of lakes in the most
disturbed condition for chemical, physical, and human use measures, e.g., what proportion of lakes have excess
phosphorus concentrations. The second examines the severity of the impact from an individual stressor when it is
present, e.g., how severe is the biological impact when excess phosphorus levels occur. Ranking ultimately requires
taking both of these perspectives into consideration. The third approach involves attributable risk, which is a value
derived by combining the first two risk values into a single number for ranking across lakes.
Throughout this section, relative and attributable risks of chemical and physical stressors to the benthic
macroinvertebrate indicator are assessed and reported as if the stressors were entirely independent. Consequently,
the attributable risk percentages do not sum to 100%. In reality, lakes typically experience multiple stresses
simultaneously, resulting in cumulative effects that cannot easily be attributed to a single stressor. Similar results for
zooplankton are available online.
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
0.
-------
6.2.1 Relative Extent of Most Disturbed Condition
Relative extent depicts the percentage of lakes in the most disturbed condition and is
a way of evaluating how pervasive a particular condition is. An indicator with a high
relative extent is widespread nationally. Conversely, indicators that occur either in a small
area or infrequently across a wide area will have lower relative extent estimates.
The NLA indicators with the most widespread occurrence are phosphorus,
methylmercury in sediments, and nitrogen, which are in the most disturbed condition in
40%, 40%, and 35% of lakes, respectively (Figure 6.1).
6.2.2 Relative Risk
The evaluation of simplified relative risk ratios is a way to estimate the severity of effects.
Relative risk conveys the likelihood of having poor biological condition when a particular
stressor is high. For example, one can examine the iikelihood of having poor biological
conditions when phosphorus concentrations are high compared with the likelihood of
poor biological conditions when phosphorus concentrations are low or moderate. When
these two likelihoods are quantified, their ratio is called the relative risk. A relative risk
value less than 1 indicates no evidence for an effect of a stressor on condition. Relative
risk values greater than 1 indicate that the stressor likely has an impact on biological
condition.
Results of the relative risk analyses for benthic macroinvertebrates are presented in
the middle panel of Figure 6.1. For benthic macroinvertebrates, total phosphorus
has the highest relative risk estimate nationally (2.2). This means that lakes with high
levels of phosphorus are about 2.2 times more iikeiy to also have most disturbed
macroinvertebrate condition. Next, lakeshore disturbance and total nitrogen show risk
values similar to one another; lakes in the most disturbed condition for these indicators
are about 1.6 times more likely to also have most disturbed macroinvertebrate condition.
6.2.3 Attributable Risk
Attributable risk represents the estimated effect of potential stressors; it allows us to
rankstressors based on the magnitude of the improvement to biological condition that
would result from reducing those stressors. It is derived by combining relative extent
and relative risk into a single number (see third panel of Figure 6.1). This risk number is
presented in terms of the percentage of lakes that could be improved - that is, moved
into either the least- or moderately disturbed condition from most disturbed - if high
levels of the stressor were reduced to low or moderate.
Attributable risk involves assumptions, including: 1) that a causa! relationship exists,
where the stressor causes a lake to have an increased probability of being in a most
disturbed biological condition; 2) that effects would be reversed if the stressor were
removed from a lake in most disturbed condition; and 3) that the stressor's impact
on a lake's biological condition is independent of other stressors. These assumptions
are difficult to meet with survey data like those collected in the NLA. Despite this
limitation, attributable risk can provide a hypothesis of which stressors might be higher
priorities at the national level. Considering attributable risk rankings can serve as a
too! for policymakers and managers when making decisions related to restoration and
protection. Estimates for attributable risk for the benthic macroinvertebrate biological
condition are presented in the right panel of Figure 6.1.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
a
-------
Figure 6.1: Estimated Risk to Biota Caused by Stressors
In relation to: Benthic Invertebrates | National
Relative Extent (% of Lakes in Most Disturbed Condition)
0% 20% 40% 60% 80%
Relative Risk
2%
Human
Use
Chemical Acidification 0%
Nitrogen (Total)
Oxygen (Dissolved)
Phosphorus (Total)
Trophic State
Atrazine (Detected)
Mercury (Methyl)
Mercury (Total)
Microcystin (Detected)
Microcystin (Risk)
Physical Lake Drawdown Exposure 6%
Lake Habitat Complexity
Lakeshore Disturbance I
Riparian Vegetation Cover
Shallow Water Habitat
35%
40%
21°
30%
40%
26%
39%
0%
29%
18%
3- 28%
18%
I 0.3
Attributable Risk
0% 20% 40% 60% 80%
QJ- 0.8
I
QM 0.8
&
0.7
0.9
1.5
1.4
1.0
Increased risk
Va ues at or be ow zero hot shown
Relative Extent
Relative extent depicts the
percentage of lakes in the
most disturbed condition.
In 2012, EPA found that
40% (45,262) of all
national lakes are
designated as most
disturbed for phosphorus
(total). The confidence
interval for this estimate is
34% to 47% (36,637 to
53,886 lakes).
Relative Risk
Relative risk conveys the
likelihood of having poor
biological condition when
a particular stressor is
high. In 2012, EPA found
that when phosphorus
(total) is present at most
disturbed levels, benthic
invertebrates are 2.2 times
more likely to be in a most
disturbed condition. The
confidence interval for this
estimate is 1.6 to 3.1.
Attributable Risk
Attributable risk is presented
in terms of the percentage
of lakes in the most
disturbed condition for
biological condition that
could be improved if high
levels of a particular stressor
were reduced to low or
moderate levels. In 2012,
EPA found that the number
of lakes in the most
disturbed condition for
benthic macroinvertebrates
could be reduced by
approximately 35% if lakes
with phosphorus (total)
levels were improved to
moderate or least disturbed
conditions. The confidence
interval associated with this
estimate is 19% to 48%.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
0l
-------
The attributable
risk results
suggest a need to
focus on reducing
nutrients in our
lakes.
Two NLA indicators have statistically significant attributable risk levels. Based
on the NLA 2012 assessment, total phosphorus has the highest attributable
risk at 35% (Figure 6.1). Phosphorus is followed by total nitrogen, which has an
attributable risk estimate of 16%.
As an example of interpreting attributable risk, phosphorus occurs at high levels
in 40% of lakes. Lakes are more than twice as likely to have poor conditions
for benthic macroinvertebrates when phosphorus is high (relative risk of 2.2).
Relative extent and relative risk, combined, result in an attributable risk level
of about 35%. That is, if phosphorus levels were reduced such that lakes were
moved out of the most disturbed category, one might expect to see 35% of lakes
move from most disturbed to least disturbed or moderate conditions for benthic
macroinvertebrates.
6.3 IMPLICATIONS FOR LAKE MANAGERS
The NLA provides a number of findings that lake managers can use to protect and
restore lakes. It is important to keep in mind, however, that while survey results fill
information gaps in national monitoring by generating estimates of the condition
of water resources, evaluating the prevalence of key stressors, and documenting
changes in the population of waters over time, they do not address all information
needs at all scales. For example, the lakes survey does not address causal factors
or sources of stress. In-depth monitoring and analysis of individual lakes, whether
already being carried out or needed in the future, are required to establish
causality and to better inform restoration.
Nutrients have been longstanding stressors of water bodies in this country.
Nationally, over 40% of lakes exhibit most disturbed conditions for phosphorus,
while 35% are in the most disturbed condition for nitrogen. Other widespread
stressors include methylmercury, where 40% of lakes are in the most disturbed
condition, and two physical habitat measures, lake habitat complexity and riparian
vegetation cover (29% and 28% are in the most disturbed condition, respectively).
Lake managers could also consider the national 2007-2012 comparison
information in evaluating site-specific information in a broader context.
Conducted on a five-year basis, future iake surveys will help water resource
managers assess broad-scale temporal differences in the data and perform trend
analyses.
States, tribes, and others may consider using NLA regional data and methods to
meet their lake assessment needs and to inform resource management priorities
and actions.
I
uy {
, <&* ¦-
wmsmHI
6.4 NEXT STEPS FOR THE NATIONAL SURVEYS
EPA, in partnership with states and tribes, produces national water quality
assessments on a regular cycle under the NARS program. With the release of the
National Wetland Condition Assessment in 2016, all waterbody types - rivers and
streams, lakes, coastal and Great Lakes waters, and wetlands - have been assessed
at least once.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
(E
-------
EPA is committed to continually enhancing and refining these surveys. As technologies and analytical methods
advance, future surveys may also include new indicators and new ways of presenting findings. In the case of the NLA,
EPA and its partners have considered the results of this second survey and discussed whether changes are needed to
the design, assessment indicators, field methods, laboratory methods, and/or analysis procedures. Sampling for the
third NLA will take place in the summer of 2017. EPA encourages researchers to further explore the findings of this and
other surveys in the series.
The NLA 2012 would not have been possible without the involvement of scientists and resource managers working
for state and tribal agencies across the United States. EPA will continue to help state and tribal partners translate the
expertise gained through these national surveys in carrying out studies of their own waters. Finally, we will work to
encourage wide use of the NLA data to evaluate the success of efforts to protect and restore the quality of U.S. waters.
We hope that NLA 2012 data are used in subsequent analyses to improve assessment results and to enhance general
ecological understanding.
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
-------
Brown, L.D., Cai, T.T. and DasGupta, A., 2001. Interval estimation for a binomial proportion. Statistical science, pp. 101 -
117.
Gilliom and others, 2006. The Quality of Our Nation's Waters—Pesticides in the Nation's Streams and Ground Water,
1992-2001: U.S. Geological Survey Circular 1291,172 p.
Stoddard, J.L., J. Van Sickle, A.T. Herlihy, J. Brahney, S. Paulsen, D.V. Peck, R. Mitchell, and A.I. Pollard, 2016. Continental-
scale increase in lake and stream phosphorus: are oligotrophic systems disappearing in the United States?
Environmental Science and Technology, 50(7):3409-3415. doi:10.1021/acs.est.5b05950.
U.S. Environmental Protection Agency (USEPA), 2009. National Lakes Assessment: A Collaborative Survey of the Nation's
Lakes. EPA 841 -R-09-001. U.S. Environmental Protection Agency, Office of Water and Office of Research and
Development, Washington, D.C.
USEPA, 2011.2012 National Lakes Assessment. Field Operations Manual. EPA 841 -B-11 -003. U.S. Environmental
Protection Agency, Washington, DC.
U.S. Environmental Protection Agency (USEPA), 2016. National Lakes Assessment:Technical Report. EPA 841 -R-16-114.
U.S. EPA, Washington, DC.
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
©
-------
Appendix A
The following lookup table provides summary information for each of the NLA 2012 indicators.
Category Indicator Benchmark approach General assessment notes
assessed?
Collected from the lake bottom at 10 shoreline locations and composited
for each lake. Organisms were usually identified to genus and an index
was developed based on life history characteristics and tolerance to
environmental conditions.
Zooplankton
NLA-derived, regionally
specific benchmarks
No
Collected from the water column at the open-water site. Organisms were
usually identified to genus and an index was developed based on life
history characteristics and tolerance to environmental conditions.
Chemical Acidification
Nationally consistent,
literature-benchmark
No
ANC (corrected for DOC) measured from a vertically integrated water
column at the open-water site. Measured concentrations were compared
to benchmarks.
Oxygen (Dissolved)
Nationally consistent,
literature- benchmark
Yes
Measures were collected from the in-situ oxygen measure from the top 2m
of the profile at the index site. Measured concentrations were compared to
benchmarks.
Nitrogen (Total)
NLA-derived, regionally
Yes
Collected from a vertically integrated water column at the open-water site.
specific benchmarks
Measured concentrations were compared to benchmarks.
Phosphorus (Total)
NLA-derived, regionally
Yes
Collected from a vertically integrated water column at the open-water site.
specific benchmarks
Measured concentrations were compared to benchmarks.
, . _ Nationally consistent, ,, A trophic state index was calculated based on measured chlorophyll-a
Trophic State , , , Yes p 1 y
literature-benchmark concentration.
, . , Benthic NLA-derived, regionally
Biological 7 Yes
macroinvertebrates specific benchmarks
National Lakes Assessment 2012 I A Collaborative Survey of Lakes in the United States
34
-------
Category
Indicator
Benchmark approach
Was change
assessed?
General assessment notes
Human Use
Atrazine
Nationally consistent,
literature-benchmark
No
Collected from a vertically integrated water column at the open-water site.
We report on detection; measured concentrations were compared to an
EPA plant-effects benchmark.
Chlorophyll-a (Risk)
Nationally consistent,
literature-benchmark
Yes
Collected from a vertically integrated water column at the open-water
site. Concentrations were compared to WHO algal toxin benchmark for
recreation.
Cyanobacteria (Risk)
Nationally consistent,
literature-benchmark
Yes
Collected from a vertically integrated water column at the open-water
site. Concentrations were compared to WHO algal toxin benchmark for
recreation.
Mercury (Methyl)
Nationally consistent,
Yes
Collected from the top 2cm of sediment from a core taken from the
literature-benchmark
bottom of the lake. Concentrations were compared to a benchmark.
Microcystin (Risk)
Nationally consistent,
literature-benchmark
Yes
Collected from a vertically integrated water column at the open-water site.
We report on detection; measured concentrations were compared to WHO
algal toxin benchmark for recreation.
Mercury (Total)
Nationally consistent,
Yes
Collected from the top 2cm of sediment from a core taken from the
literature-benchmark
bottom of the lake. Concentrations were compared to a benchmark.
Physical
Lake drawdown
NLA-derived, regionally
Yes
Observations were recorded from 10 shoreline locations around each lake.
exposure
specific benchmarks
Information was compared to regionally specific benchmarks.
Lake habitat
NLA-derived, regionally
Yes
Observations were recorded from 10 shoreline locations around each lake.
complexity
specific benchmarks
Information was compared to regionally specific benchmarks.
Lakeshore disturbance
Nationally consistent,
Yes
Observations were recorded from 10 shoreline locations around each lake.
literature-benchmark
Information was compared to regionally specific benchmarks.
Riparian vegetation
NLA-derived, regionally
Yes
Observations were recorded from 10 shoreline locations around each lake.
cover
specific benchmarks
Information was compared to regionally specific benchmarks.
Shallow water habitat
NLA-derived, regionally
Yes
Observations were recorded from 10 shoreline locations around each lake.
specific benchmarks
Information was compared to regionally specific benchmarks.
National Lakes Assessment 201 2 I A Collaborative Survey of Lakes in the United States
0-
-------
NLA 2012 Report Photographs
Report page number
Photograph location
Photograph courtesy of
Cover and section header (left
3 and right)
Eric Vance, EPA
Cover (2nd to right) and
section header
Lake Wood, Maine
Hilary Snook, EPA Region 1
1
Katie DeGoosh, Rhode Island Department of Environmental Management
2
Eric Vance, EPA
Section 2 header (left)
Kansas
Debbie Baker, Kansas Biological Survey
3
Long Pond, Massachusetts
Susan Holdsworth, EPA
7
Eric Vance, EPA
8
Lake Pleasant, New York
Susan Holdsworth, EPA
13
Eric Vance, EPA
14
Sacheen Lake, Washington
Jenny Hall, Washington Department of Ecology
16
Anne Rogers,Texas Parks and Wildlife Department
17 (left)
Virginia
EPA
17 (2nd from left)
Saguaro Lake, Arizona
EPA
17 (mid)
Pyramid Lake, Nevada
Marianne Denton, Nevada Division of Environmental Protection
17 (2nd from right)
Elaine Lake, Tennessee
Tetra Tech
17 (right)
Swofford Pond, Washington
Jenny Hall, Washington Department of Ecology
18
Wilmington, North Carolina
Compiled by National Environmental Education Foundation for USEPA
19
Kansas
Debbie Baker, Kansas Biological Survey
21
Peace Dale Reservoir, Rhode Island
Katie DeGoosh, Rhode Island Department of Environmental Management
21
Slatersville Reservoir, Rhode Island
Katie DeGoosh, Rhode Island Department of Environmental Management
23
Belleville Pond, Rhode Island
Katie DeGoosh, Rhode Island Department of Environmental Management
24
Belleville Pond, Rhode Island
Katie DeGoosh, Rhode Island Department of Environmental Management
26
Missouri
Tony Thorpe, University of Missouri
27
Wisconsin
https://pixabay.com/en/wisconsin-lake-water-sailboat-140392/
29 (top)
Eric Vance, EPA
29 (2nd down)
Belleville Pond, Rhode Island
Katie DeGoosh, Rhode Island Department of Environmental Management
29 (3rd down)
Oklahoma
Jessie Stine, Oklahoma Water Resources Board
29 (4th down)
Baker Lake, Washington
Jenny Hall, Washington Department of Ecology
29 (5th down)
https://pixabay.com/en/duck-drake-mallard-plumage-711443/
29 (bottom)
Long Pond, Massachusetts
Susan Holdsworth, EPA
31
Eric Vance, EPA
34 (left)
Spring Lake, Vermont
Hilary Snook, EPA Region 1
34 (2nd from left)
Swiggetts Pond, Delaware
Ellen Dickey, Delaware Division of Water Resources
34 (3rd from left)
Kansas
Debbie Baker, Kansas Biological Survey
------- |