Ecological Condition of
Lakes in Idaho, Oregon,
and Washington
-------
-------
Ecological Condition of Lakes in Idaho, Oregon, and Washington
EPA Region 10 Report
Authors:
L.G. Merger, P.T. Leinenbach, and G.A. Hayslip
December 2010
U.S. Environmental Protection Agency, Region 10
Office of Environmental Assessment
1200 Sixth Avenue, Suite 900
Seattle, Washington 98101
Publication Number: EPA-910-R-10-001
Suggested Citation:
Merger, L.G, P.T. Leinenbach, and G.A. Hayslip. 2011. Ecological Condition of Lakes
in Idaho, Oregon, and Washington. EPA-910-R-10-001. U.S. Environmental Protection
Agency, Region 10, Seattle, Washington.
This document is available at: http://yosemite.epa.gov/r10/OEA.NSF/Monitoring/Lakes/
-------
[Blank Page]
-------
Region 10 Lakes Assessment
Table of Contents
Acknowledgements iv
Abstract iv
Purpose 1
Introduction 1
Overview of Lakes Ecological Assessment 2
A. Survey Design 2
B. Lake Selection and Extent of Evaluation 3
C. Description of Region 10 Sample Lakes 4
D. Aquatic Stressor Indicators 7
E. Landscape Indicators 9
Assessment Thresholds 9
Numeric Thresholds 9
Reference Condition 9
Methods 10
A. Quality Assurance 10
B. Field Sample Collection 11
C. Landscape Data Methods 12
1. Biological Condition 14
2. Water Quality and Trophic State 16
A. Water Quality 16
B. Trophic State 19
3. Physical Habitat Stressors 21
4. Suitability for Recreation 24
A. Algal toxins 24
B. Pathogens 26
C. Contaminants in Fish Tissue 27
Summary of Findings: ranking stressors 27
Recommendations 29
References 31
Appendix 1. Sampled Lakes in Region 10 33
Appendix 2. Analysis of Landscape Metrics 36
Introduction 36
Objectives 36
Identifying Best Metrics 36
Methods 36
Results 37
Indicator Development 40
Methods 40
Results 43
Future Work 43
Appendix 3. Additional Nutrient Analysis for Region 10 NLA Lakes 48
Appendix 4. Fish Tissue—Region 10 results from 2004 National Lakes Fish Tissue
Survey 58
-------
Region 10 Lakes Assessment
List of Figures
Figure 1. Fate of lakes available from NHD for use as target sample. Lake criteria
requirements and screening for sampleablity yields inference population of 1,700 lakes.
4
Figure 2. Description of sampled lakes by aggregated Level 3 Ecoregion and by
Ownership 5
Figure 3. Sample lakes in each size category (N=90) 6
Figure 4. Proportion of inference lakes (1700) within each lake origin category
compared to proportion when weighting is not applied (90 sampled lakes) 7
Figure 5. Diagram of sampling locations within each lake 11
Figure 6. Example of lake contributing area with sample cover classes and illustration
of four buffers distances, entire watershed, 5km, 2km and 200m, in which each cover
attribute was calculated (Smith reservoir, Oregon) 13
Figure 7. Region 10 lakes in good, fair, poor, no data, and not assessed condition
classes for the plankton community 0/E and sediment lake diatom condition (LDC)
indicators of biological condition 16
Figure 8. Lakes in good, fair, poor condition classes for total phosphorus and nitrogen.
18
Figure 9. Lakes in good, fair, poor condition classes for turbidity 19
Figure 10. Percentage of inference lakes in each of four trophic categories based on 20
Figure 11. Percent of lakes in four trophic classes based on TSI thresholds for
chlorophyll-a 21
Figure 12. Lakes in good, fair, poor condition classes for lakeshore habitat and shallow
water habitat 23
Figure 13. Lakes in good, fair, poor condition classes for habitat complexity and
lakeshore disturbance 24
Figure 14. Lakes within low, medium and high risk categories for algal toxin exposure
based on cyanobacteria cell counts and chlorophyll-a concentration 26
Figure 15. Relative extent of poor stressor condition across all stressor metrics for all
Region 10 lakes 28
Figure 16. Relative extent of poor stressor condition across all stressor metrics for
natural lakes and man-made lake categories 29
-------
Region 10 Lakes Assessment
List of Tables
Table 1. Description of Indicators used to evaluate lake condition 8
Table 2. Summary of field data collection protocols 12
Table 3. Summary of landscape metrics used for comparison to lake condition 14
Table 4. Water quality summary statistics for Region 10 lakes' inference population
(-1700 lakes) 17
Table 5. Carlson Trophic State Index (TSI) parameter thresholds 20
Table 6. World Health Organization's recreation indicator thresholds of risk associated
with potential exposure to algal toxins 25
Table 7. Results of ranking Region 10 data by Enterococcus concentration categories.
Samples collected from 88 sites (inference population =1694) 26
List of Maps
Map 1. Locations of sampled lakes in the Coastal/Mountain 5
-------
Region 10 Lakes Assessment
Acknowledgements
This project was completed with the help of numerous individuals and agencies. EPA's
Office of Water was responsible for data management including quality assurance,
metric calculations, data analyses, and generation of condition class indicators. This
work was accomplished collaboratively with EPA's Office of Research and Development
(ORD), Corvallis, OR, and scientists from other agencies and organizations. ORD
scientists, Phil Kaufmann, David Peck, John Van Sickle, and Tony Olsen, provided
advice on this reporting of the Region 10 data. We also thank Alan Herlihy and Thorn
Whittier, Oregon State University, for their helpful advice. Field data in Region 10 were
collected by state environmental monitoring departments: Idaho DEQ, Oregon DEQ,
and Washington Ecology. Jason Pappani and Mary Anne Kostermann, Idaho
Department of Environmental Quality, Shannon Hubler, Oregon Department of
Environmental Quality, and Maggie Bell-McKinnon, Washington Department of Ecology
were our state collaborators and were critical to the success of the Region 10 portion of
this national survey. Scott Fields (Coeur d'Alene Tribe) and Maggie Bell-McKinnon
contributed cover photos.
Abstract
The Environmental Protection Agency (EPA) in collaboration with state agencies
conducts monitoring of various aquatic resources to answer questions on the condition
of the Nation's waters. This series of surveys is conducted under EPA's National
Aquatic Resources Surveys (NARS) program. These surveys employ a statistical
design that makes it possible to describe the quality of the resource across the Nation in
terms of good, fair or poor condition relative to a reference condition or numeric
standards. In addition, the extent of human disturbance can be described in relation to
geographic variations.
The NARS program initiated a survey of lakes called the National Lakes Assessment
(NLA). Data were collected for this assessment from 1,028 lakes across the contiguous
United States in summer 2007. This report uses a subset of the data from this larger
project to assess the ecological condition of the lakes of the EPA Region 10 states of
Idaho, Oregon, and Washington (subsequently referred to as Region 10). A total of 90
lakes were sampled in Region 10. In general, most lakes in the Region are in good or
fair condition based on the results of the indicators analyzed. The most widespread
stressors are physical habitat quality of the lakeshore and shallow areas, and nutrients.
IV
-------
Region 10 Lakes Assessment
Purpose
Lakes are an important aquatic resource in the Pacific Northwest Region. Monitoring of
lakes by the states within EPA Region 10 is limited and there are no programs that
survey lakes for the purpose of state-wide assessment. This ecological assessment of
lakes in Region 10 has three purposes:
• Report on the ecological condition of lakes using direct measures of biological
assemblages.
• Identify and rank the relative importance of stressors affecting lake condition by
using measures of chemical, physical and biological habitat to determine how
wide-spread/common these stressors are.
• Conduct preliminary analysis of landscape metrics derived from remotely sensed
data that may be useful for assessing regional lake condition.
Introduction
The EPA conducts nation-wide ecological surveys of aquatic resources to evaluate their
status and to examine associations between ecological condition and natural and
anthropogenic influences. These National Aquatic Resource Surveys (NARS) are based
on the premise that the condition of aquatic biota can be determined by examining
biological response indicators and ecological indicators of stress. The long-term goal of
NARS is to directly measure environmental resources to determine if they are in an
acceptable or unacceptable condition relative to a set of environmental or ecological
values. Two major features of these surveys are the use of ecological indicators and
probability-based selection of sample sites.
An ecological assessment can be performed in a variety of ways ranging from a
description of the extent of a resource to an enumeration of the abundance and
distribution of biota in an ecosystem. The ecological assessment of lakes in Region 10
described in this report evaluates critical stressors related to water quality, biological
condition, physical habitat condition, and recreation. Two critical components of aquatic
ecosystems are: 1) the condition of the biota, and 2) the relative importance of human-
caused stressors.
The first component of this ecological assessment is based on the fact that biological
communities are adapted to local habitat (the combination of physical, chemical, and
spatial elements) and therefore the ecological condition of lakes is reflected by the
quality and health of the biotic communities. Essentially, the biotic communities
integrate the many human disturbances that we are interested in assessing.
Maintaining the biotic communities is also one of the pillars of the federal Clean Water
Act ".... [Supporting and maintaining a balanced, integrated, adaptive community of
1
-------
Region 10 Lakes Assessment
organisms having a species composition, diversity, and functional organization
comparable to that of the natural habitat."
The second component of this assessment evaluates ecological stressors. Stressors
are defined as the pressures or disturbances exerted on aquatic systems. These are
the chemical, physical, and biological components of the ecosystem that have the
potential to degrade the biotic integrity of the aquatic system. This ecological
assessment will identify stressors and describe their spatial extent.
The National Lakes Assessment (NLA) was a two-year effort to collect data from lakes
across the United States and to report on their ecological condition at a national scale
(USEPA 2010). Consistent field, lab, and data analysis methods were used across the
country and across lake types. All sites were selected using a probabilistic design.
Collectively, the sites are a statistical representation of the target population of lakes of
the United States.
This Region 10 report uses a subset of the data analyzed for the NLA to describe the
condition of the lakes in EPA Region 10. Region 10 has previously reported the
ecological condition of other aquatic resources including rivers, streams and estuaries in
Region 10 using this same approach (Hayslip et al. 2006, Merger et. al. 2007). This is
the first time extensive data have been available for reporting on the ecological
condition of lakes across Region 10.
Overview of Lakes Ecological Assessment
A. Survey Design
Assessing a very large and diverse area requires a study design that can adequately
capture the variation across the landscape and be descriptive of the entire ecological
resource of lakes in the Region. There are various options for collecting data to
describe the ecological condition of this 'target population'. A census method, where
data are collected from every lake, is impractical (if not impossible). This survey used a
sample approach similar to a public opinion poll where data are collected from a subset
of the target lakes. This information is then used to determine characteristics of the
greater target population. The sample sites are selected using a probability-based
sampling method to insure that they are statistically representative of the target
population. In a probability sample, every lake of the target population has a known
non-zero probability of being selected. This feature has two advantages in that it 1)
prevents site selection bias and 2) allows statistically valid inferences describing
characteristics of the entire target population to be made.
The target population was sampled in a spatially-restricted manner so that the
distribution of the sample sites has approximately the same spatial distribution as the
target population. This was achieved by using an unequal probability sample method to
ensure distribution of samples of sites by size, State, and major ecoregion types
-------
Region 10 Lakes Assessment
(mountainous/humid v. xeric). For example, in this study large lakes were given higher
probability of being selected for sampling than small lakes. This effectively increases
the probability of having large lakes selected for the sample so that the sample is not
dominated by the small lakes, which are much more common in the landscape. This
variable selection probability by lake size is accounted for when making the regional
estimates by using site weighting factors. Each site is assigned a weight, based on the
occurrence of its type in the Region. Small lakes have a smaller weighting factor than
large lakes. Therefore, any inferences based on the un-weighted set of sites to the
entire target population are inaccurate.
B. Lake Selection and Extent of Evaluation
The sample frame for this study was all lakes, reservoirs, and ponds that are permanent
waterbodies within Oregon, Idaho, and Washington that have the following
characteristics:
• surface area > 4 ha (10 acres)
• > 1000m2 open, unvegetated, water surface area
• > 1 meter depth
• excludes 'working' lakes meaning those used for aquaculture, tailings disposal,
active borrow pits, sewage treatment, evaporation, etc.
• excludes saline coastal waterbodies and those under tidal influence
• excludes flowing waterbodies such as 'run of the river' reservoirs
The data used to generate this set of lakes was the National Hydrography Dataset
(NHD), which is a set of CIS layers. For Region 10, this dataset consisted of 11,340
lakes, reservoirs and ponds with at least four hectares of surface area. This dataset
was screened using the criteria described above which resulted in a set of 3,423 lakes
considered the target population for Region 10.
Sample lakes were selected randomly from the target population in proportion to their
occurrence within five size categories (Overton et al. 1990, Stevens and Olsen 2004).
A final evaluation was conducted to ensure each lake was accessible and satisfied the
criteria for inclusion as a target lake. For this project, accessible meant that 1) they did
not have safety issues that would prohibit sampling, 2) they were not excessively
remote, and 3) that landowners would grant access.
In Region 10, a total of 90 lakes were slated for sampling. As lakes were selected from
the design list of eligible lakes (those meeting the target criteria), many were rejected
due to access denial or inaccessibility. Almost 50% of the eligible lakes were rejected
before the final 90 lakes were selected. These 90 lakes represent the "inference
population" of 1,700 (49.7 % of the original target population). Results in this report are
presented in relation to this inference population. For example, if an indicator has poor
condition in 10% of the lakes this means that 170 lakes in the Region are in poor
condition (Figure 1).
-------
Region 10 Lakes Assessment
Target Population = 3,423 lakes
Target sampled = 1,700
Inference Population = 1,700 lakes
Sampleable = 1,700
Figure 1. Fate of lakes available from NHD for use as target sample. Lake criteria requirements and
screening for sampleablity yields inference population of 1,700 lakes.
The process used to select sites allows for unbiased estimates of condition indicators
with known statistical confidence. Error bars on the graphs in this report express the
uncertainty of the population estimates at the 95% confidence level. For example, an
estimate of poor biological condition of 40% with an error bar of +/-10% means there is
95% confidence that the true value is between 30% and 50%.
Error bars of 5-15% would be expected for this type of ecological assessment.
However, In Region 10, error bars tended to be rather large (approximately 20%)
depending on the indicator. One explanation for the large error bars in the Region is the
heterogeneous landscape and lack of consistent lake size in the Region. Further, the
variance is calculated spatially meaning that variances are all based on each site and its
three closest "neighbors" in geographic space. If sites have neighbors with large site
weights, meaning they are rare in terms of size (big lakes), then larger local variance
estimates will result. If there are areas with no or very few lakes, nearest neighbors
may be far apart, or at different elevations. The spatial variability in distance between
lakes, elevation, and variation in lake size results in the increased error in the region
compared to areas with more homogeneous lake size by elevation (e.g. Minnesota,
Wisconsin) and compared to the nation-wide assessment error (USEPA 2009a).
C. Description of Region 10 Sample Lakes
The Pacific Northwest ecosystem has diverse physiological, climatic, and floral and
faunal characteristics as evident by the inclusion of all or portions of 11 different
ecological regions (ecoregions) within its boundary (USEPA 2003, Omernik 1987). The
diversity of the contiguous portion of Region 10 includes large expanses of high dry
plateau, steep mountains, and forested areas. The Region has two major climatic
regions, xeric and mountainous areas. The xeric portion of the basin is represented by
the Columbia Plateau, North Basin Range, and Snake River Plain ecoregions. The land
-------
Region 10 Lakes Assessment
area categorized as xeric is a substantial portion of the PNW (shown in brown on Map
1). The remaining ecoregions comprise the mountainous climatic region (shown in
green).
. •*;
^Probability
^Hand-picked
Map 1. Locations of sampled lakes in the Coastal/Mountain
west (green) and Xeric (brown) ecoregions in Region 10 states
of Oregon, Washington, and Oregon.
The location of the 90 randomly selected sample lakes used to represent the inference
population of 1,700 lakes are shown on Map 1 and are listed in Appendix 1. Nearly
three quarters of the sampled lakes are within the coastal/mountains ecoregions, and
most are in public ownership (Figure 2).
private
18%
tribal/public
1%
public
75%
Figure 2. Description of sampled lakes by aggregated Level 3 Ecoregion and by Ownership.
-------
Region 10 Lakes Assessment
Lakes from five size categories of open water surface area were included in the sample.
The 90 sample lakes were relatively equally distributed between the three mid-sized
categories (Figure 3).
4-10 10-20 20-50 50-100 >100
surface area (ha)
Figure 3. Sample lakes in each size category (N=90).
Each lake sampled in the smallest size class represents a proportionately larger number
of lakes in the PNW states due to the far larger number of small lakes present in the
three-state region relative to larger size classes. Thus, these small lakes carry a larger
'weight' in the survey. The sample design forced the inclusion of large sized lakes in
the sample. Because of the proportionately larger number of small lakes a completely
random sample without this 'forcing' would result in a dominance of small lakes in the
sample, making it difficult to draw inferences for large lakes.
Applying the assigned weights to the sample lakes results alters the proportion of lakes
within the man-made versus natural categories that will be represented by the
assessment results. The inference population has 33% of lakes in the man-made
categories (including reservoirs), versus 43% in the 90 lakes actually sampled (Figure
4). Similarly the proportion of lake distribution among ecoregions differs once weights
are applied: about 8% (114 lakes) of the inference lakes are within the xeric portions of
Region 10 compared to 28% of the 90 sampled lakes.
-------
Region 10 Lakes Assessment
Reservoir
7%
Man-
made
20%
Natural
67%
Reservoir
23%
Natural
57%
Figure 4. Proportion of inference lakes (1700) within each lake origin category compared to proportion
when weighting is not applied (90 sampled lakes).
D. Aquatic Stressor Indicators
Characterizing the ecological condition of lakes is complex due to their dynamic nature.
Ecological indicators must be carefully selected so that they robustly represent the
important aspects of lake quality. In order to thoroughly characterize lakes, indicators
were selected to assess four major aspects of lake ecological characteristics. These are
biological condition, water quality and trophic state, physical habitat condition, and
recreation suitability. This report examines the most relevant metrics for evaluating the
stressors affecting the ecological condition. The four categories of lake condition
indicators and the specific indicators examined in this assessment are summarized in
Table!
-------
Region 10 Lakes Assessment
Table 1. Description of Indicators used to evaluate lake condition.
Indicator
Why important
Metrics of assessment
1. Biological Condition
Plankton: floating microscopic rganisms
includes: zooplankton ( invertebrates)
and Phytoplankton- algae
Sediment diatoms — microscopic algae
with silicon cell walls that are preserved
in lake sediments
Responds to stressors of nutrient
enrichment and turbidity. Responses
can be quantified through changes in
species composition and abundance.
Diatom species have specific
requirements of alkalinity, total P,
conductivity, etc.
Taxa loss (O/E) model
develop for the combined
presence of zooplankton
and phytoplankton
Diatom condition index
(LDC) based on diatom
assemblage in surface
sediments
2. Water Quality and Trophic State
Nutrients: total phosphorus (P) and total
nitrogen (N)
Dissolved oxygen
Water clarity-Seech i transparency depth
and turbidity measurement
Chlorophyll-a- photosynthesizing
pigment is a measurement of algal
biomass
P and N are required for primary
productivity. Excessive nutrients
negatively affect lake function.
Essential for support of aquatic life.
Organisms have differing
requirements for optimal growth.
Water clarity is an indirect measure
of algal growth and suspended
solids.
Indicates level of primary productivity
and is used to estimate trophic State.
Reference condition
comparisons and numeric
thresholds
Numeric thresholds from
the literature
Reference condition
comparisons and numeric
thresholds.
Comparison to reference
condition
3. Physical Habitat Condition
Littoral habitat — forms the interface
between terrestrial and aquatic
environment. Metric of cover features
Lakeshore habitat (terrestrial nearshore
area and shoreline zone)-metric of
vegetation structure and complexity
Physical habitat complexity — combines
littoral and lakeshore habitat metrics
Human disturbance- extent of human
activities in proximity to the lakeshore
Important area of nutrient inputs and
aquatic habitat. Area commonly
altered by human activities.
Area where human disturbance can
have the most effect on a lake.
Used to assess overall habitat
structural complexity and integrity.
Alteration of habitat by humans can
affect biological integrity.
Comparison to reference
condition
Comparison to reference
condition
Comparison to reference
condition
Comparison to reference
condition
4. Recreation Suitability
Cyanobacteria cell counts-class of
algae that produce algal toxins
Chlorophyll-a density- indicator of
presence of all algae
Microcystin -an algal toxin produced by
some cyanobacteria
Enterococci-bacteria of animal intestinal
tracts (including humans)
Fish tissue contaminants
Indirect indicator of presence of
algae that could produce toxins.
Indirect indicator of presence of
algae that could produce toxins.
Direct measure of microcystin which
is harmful to humans and wildlife.
Direct measure of bacteria.
Bioaccumulative and persistent -
these are harmful to fish consumers.
Numeric thresholds from
the literature
Numeric thresholds from
the literature
Numeric thresholds from
the literature
Relative ranking
Numeric thresholds from
the literature
8
-------
Region 10 Lakes Assessment
E. Landscape Indicators
Landscape metrics that describe the physical conditions of watersheds were calculated
for the watersheds of each lake using a Geographic Information System (CIS). These
indicators are useful for describing the type and quantity of human disturbance that can
influence lake condition. We analyzed these indicators to explore which ones are most
related to stressor indicators generated from the lakes data and to examine the buffer
widths that were most useful for these various types of landscape indicators. This is an
exploratory analysis of Region 10 data as landscape metrics were not used in the
National Lakes Survey.
Assessment Thresholds
Numeric Thresholds
For some indicators, values in the literature were available to evaluate lake condition.
These thresholds are well established and are widely used and accepted. Dissolved
oxygen levels were compared to EPA national recommendations. Lake trophic state
was determined by comparing lake water quality metrics to the Trophic State Index
developed by Carlson (1977 and 1983). Finally, the survey results for recreation
indicators are compared to recommendations from the World Health Organizations
(WHO) that are used to rate lake quality.
Reference Condition
Numeric condition threshold values were either not available or applicable for
determining many of the biological and physical habitat conditions, and some of the
water quality conditions for this lakes survey. In order to describe the ecological
condition of the lakes of Region 10, we must have an expectation of the ecological
condition in a relatively 'undisturbed' state. This benchmark for determining ecological
condition is commonly referred to as the 'reference condition'. A reference condition
can have many meanings. For instance, it could mean a 'pre-settlement condition', a
'desired condition', or an 'acceptable current condition' that implies some level of human
disturbance. Setting reasonable expectations for each of the indicators of ecological
condition is therefore a challenge. For this assessment, reference condition is
developed from the analysis of carefully selected sites that represent the best attainable
or 'least disturbed' watershed condition, habitat structure, water quality and biological
parameters (Hughes 1995, Stoddard et al. 2006). Deviation from the reference
condition is a measure of the effect of stressors on the ecosystem. A site is considered
to be in 'good' condition if it is in the condition we would expect to see if it were
minimally exposed to the stressors of concern (i.e., if it is equivalent to reference
condition). Thus, 'good' condition is defined relative to our expectations for a particular
system rather that against an absolute benchmark of ecosystem attributes (Bailey et al.
2004).
-------
Region 10 Lakes Assessment
The diversity of physical, chemical, and biological characteristics of the lakes of the
Pacific Northwest states must be considered when defining reference condition and
calculating lake ecological condition. For example, a lake with finer-sized substrate and
low shoreline vegetative structure may be typical of an undisturbed lake in one
ecoregion while those same characteristics may represent a more disturbed condition in
a forested/mountainous ecoregion. Because ecoregions have similar characteristics in
terms of soil, climate, geology, and vegetation, it follows that the lakes of an ecoregion
would have similar stressors as well as similar responses to those stressors. Thus,
ecoregions provide a template for refining the expected condition of lakes throughout a
broad and variable area.
Reference sites came from two sources: 1) a set of lakes that were handpicked for
sampling based on input from the state resource managers and 2) a set of lakes from
within the greater National Lakes Survey probability lakes dataset. The handpicked
sites were selected as being lakes in a minimally disturbed condition based on best
professional judgment of state lake survey coordinators. In Region 10, 11 of these
handpicked lakes were sampled in the three states (see Map 1). After data were
collected from all probability and hand-picked sites, the dataset was screened to select
sites in a least disturbed condition based on landscape data interpretation. This subset
of lakes was further screened to include only those where the water quality was
considered to be in good condition based on phosphorus, nitrogen, chloride, and sulfate
concentrations. From this two step process a subset of reference lakes were identified
from the probability dataset.
The reference lakes were used by EPA's Office of Water and Office of Research and
Development (ORD) to generate condition metrics for the various indicators. Technical
details on the development of each of these condition indicators will not be discussed
here but can be found in EPA's Technical Appendix (EPA 2010). The Technical
Appendix details the data analysis and development of metrics and condition indicators
that are reported in this assessment.
Methods
A. Quality Assurance
All data collected and generated for this report followed the Quality Assurance Project
Plan (QAPP) developed for the NLA (USEPA 2009b). The QAPP addresses all levels
of the program, from collection of field data and samples and the laboratory processing
of samples to standardized/centralized data management. Numerous crews conducted
field sampling. Consistency and adherence to the field protocols (USEPA 2007) was
insured by crews participating in training sessions, and field audits conducted by EPA
Region 10 personnel early in the field season. Also, 10% of the sites were re-sampled
to provide estimates of the variability of the metrics.
10
-------
Region 10 Lakes Assessment
B. Field Sample Collection
Field data were collected during the summer growing period ('index period'). In Region
10, the sites were sampled between June and August. Field sampling was conducted at
both the deepest point in the lake and at stations along the shoreline (Figure 5). The
deepest location, termed the X-site, is where most indicators were collected including
water chemistry, chlorophyll-a, phytoplankton, zooplankton, sediment diatoms, and algal
toxins samples. Ten 'physical habitat' stations were established along the shoreline
equidistant from one another with the location of the first site selected randomly.
Physical habitat and benthic macro-invertebrates were collected at each of the ten
stations and the enterococci sample was collected at the last station. These physical
habitat stations were not established at large lakes (greater than 5,000 hectares) and
physical habitat data and benthic invertebrates were not collected at those water
bodies. The enterococci sample was collected near the boat launch site for these big
lakes. A summary of field sampling protocols is shown in Table 2 and further details on
field methods and sample preservation and handling can be found in the field
operations manual (EPA 2007).
Littoral zone - benthic sampling area
Sub-littoral zone
Profundal zone
Observation station
positioned 10m
offshore for sampling
Index site
deepest point - chosen using
bathymetric map and/or sonar
X
•Water chemistry
•Chlorophyll a
•Phytoplankton
•Zooplankton
•Sediment diatoms
•Algal toxins
All stations equidistanl
from one another
Physical habitat and benthic
sampling stations (A-J) -
Starting point randomly
selected a priori
Habitat and benthic sampling station
15m
Shoreline >
zone (1 m)
Benthic sample collected
from dominant habitat within
littoral zone
^
•#•
0
Riparian
zone
Littoral
zone
Ll5m
MOm
bservation station
Figure 5. Diagram of sampling locations within each lake.
11
-------
Region 10 Lakes Assessment
Table 2. Summary of field data collection protocols.
Metric Type
Field Method
Sampling atX-site (deepest part of lake)
Secchi disc
transparency
Water chemistry
Water quality profile
Chlorophyll-a
Phytoplankton
Zooplankton
Sediment diatoms
Algal toxin
Sediment mercury*
Deployed from shady side of boat. Depth of disappearance and
reappearance were recorded. Secchi reading was used to define the
euphotic zone for the water chemistry data collection procedures (2 times
Secchi depth = total euphotic zone).
Water sample collected with an integrated sampler so that water could be
collected from the upper two meters of the water column, which
encompasses the euphotic zone. When Secchi depth determined that the
euphotic zone was < 2 meters the integrated sampler was only deployed to
that depth.
In situ DO, pH, water temperature, and conductivity were measured with an
electronic meter at the surface and through the water column (DO value
used in assessment was the mean value from the top 2 m).
Collected as part of water sample using the integrated sampler. Sample was
filtered in the field and filter was submitted for analysis.
Collected as part of water sample using the integrated sampler.
Conducted two vertical sampling tows through the entire water column, one
using fine (80um) and one using coarse mesh (243um) Wisconsin nets.
A 4in. diameter coring device was deployed to collect a sediment core with
minimum length of 45 cm. A one-cm slice was removed from both the top
and bottom of the core.
Collected as part of water sample using the integrated sampler.
Small plug of sediment from surface portion of sediment core was removed
with a pipette prior to removal of the 1-cm slice for the diatom sample.
Sampling at 10 physical habitat stations
Physical habitat
Benthic Invertebrates*
Enterococci
Recorded three types of visual (qualitative) observations made within the
habitat station diagramed in Figure 5:
1) Littoral habitat cover and structure recorded from 10x1 5m littoral plot
2) Riparian/shoreline vegetative structure and cover complexity at three
levels (tree canopy, understory, and ground cover) recorded from
Riparian/shoreline plot.
3) Human influences within the riparian/shoreline/littoral portions were
recorded.
A 500 urn D-frame kick net was swept through a single 1 linear meter of the
dominant habitat type at a maximum depth of 0.5m. Samples from all ten
stations were composited into one sample.
Water sample collected in littoral zone of last physical habitat station where
lake depth is one meter deep. Sample collected at 0.3 meters depth.
*=data not available for report
C. Landscape Data Methods
Basin area was delineated for each lake by EPA Office of Water using components of
the NHDPIus system. The NHD flowline data was examined to determine the lowest
Hydrologic sequence number. This number was selected as the lake outlet and its
downstream measure was used as the point location of the pour point. This point and
the flowline data associated with the waterbody was used to delineate the drainage
basin using the NHDPIus Basin Delineation tool. A few lakes that were not found on the
12
-------
Region 10 Lakes Assessment
NHDPIus Basin network had their basins delineated using the ArcGIS Spatial Analyst-
>Hydrology->Watershed tool. A sample watershed is in Figure 6.
Figure 6. Example of lake contributing area with sample cover classes and illustration of four buffers
distances, entire watershed, 5km, 2km and 200m, in which each cover attribute was calculated (Smith
reservoir, Oregon).
Within each watershed polygon, landcover metrics such as percent forest and percent
agriculture were calculated from National Land Cover Database (NLCD) Digital
coverages. The Analytical Tools Interface for Landscape Assessments (ATtlLA 3.x), an
ArcView Software extension (Ebert et al. 2000), was used to calculate the metrics. In
addition to these standard landcover metrics, EPA Region 10 generated a suite of other
metrics that are relevant to human disturbance in the Pacific Northwest. These metrics
relied on the use of higher resolution data available for the region and incorporated
several models specific to the Western US. Landscape metrics used in this analysis are
summarized in (Table 3). Additionally, landcover metrics were-calculated at three
buffer distances from the lake, 5km, 2km, and nearshore 200m (Figure 6). These are
used in the analysis to test for the optimum buffer that gives the best expression of
landscape stressor metrics. Results of the analysis of these metrics are in Appendix 2.
13
-------
Region 10 Lakes Assessment
Table 3. Summary of landscape metrics used for comparison to lake condition.
Metric Category
Land cover
Forest
disturbance
Agriculture on >
9% slope
Potential Unit
Grazing
Sediment delivery
model
Roads
Human population
Land ownership
Elevation
Precipitation
Metric Description
Percent cover of land types (% forest, shrub, agriculture, etc.)
Attributes of "Transitional from Forest harvest' and
Transitional from Forest Fire' are more detailed forest
vegetation conditions over the standard NLCD metrics which
commonly interpret 'disturbed' forest harvest as either Grass
or Shrub cover.
Amount of agriculture on steep slopes. Calculated from slope
grid, 9% threshold value, and agriculture coverage.
Indicator of the intensity of potential cattle/calf grazing.
Calculated from county cattle census data, Potential Cow
Habitat score which is in turn calculated from five inputs: land
ownership, land cover, proximity to water, topographic
position grid index, and slope.
Estimate of basin average rates of soil erosion and total
annual sediment delivery at the basin outlet. Calculated from
empirical models 'RUSLE (Renard et al. 1997) and SEDMOD
(Fraser1999).
Road density and road stream crossing density from Tela-
Atlas data layer
Population density calculated from 2000 Census Tiger File
census layer.
Four attributes of ownership: Private, State, Federal, and
Tribal.
Mean, minimum and maximum elevation of watershed.
Mean, minimum, and maximum precipitation calculated from
1 8-year dataset.
Source
NLCD
NW_GAP database
NHD and 30m NED
slope database
NLCD, NHD and 30m
NED slope database
CIS based modeling
tool from USEPA
Landscape Research
Group (Van Remortel
et al. 2005, 2006)
Region 10 SDE
database
Region 10 SDE
database
ICEBEMP website
USGS National
Elevation Dataset
(NED)
Daymet website
1. Biological Condition
Lakes have many levels of interacting biological assemblages including primary
producers such as algae and phytoplankton, intermediate assemblages such as benthic
macro- invertebrates and zooplankton, and fish assemblages operating at various
trophic levels (e.g. bottom-dwelling herbivores or pelagic predators). The overall
condition of a lake is defined by its functioning biological community and the number
and kinds of organisms in a lake is a direct measure of its health. The biotic community
determines the response indicators because alterations and disturbances to lakes
change the biotic community. Two biotic assemblages are evaluated in the survey,
plankton assemblage and sediment diatoms. Although benthic macro-invertebrates
were collected at the shore plots, the model of biological condition resulting from these
data has not been finalized by EPA Office of Water and is not presented in this report.
14
-------
Region 10 Lakes Assessment
Phytoplankton (floating algae) and zooplankton (free floating animals) are highly
responsive to changes in lake ecosystems. They are useful response indicators
because the effect of environmental changes can be detected through changes in
species composition, abundance, and body size (for zooplankton). For the NLA, project
researchers developed a combined phytoplankton-zooplankton "Observed versus
Expected Index" (0/E Index). This type of index estimates biological condition by
measuring the agreement between the taxonomic composition expected under
reference conditions and that observed at the individual lakes and thus expresses taxa
loss. The model is complicated by the fact that taxonomic composition varies with
natural environmental factors. The 0/E Index depends on models that predict how
taxonomic composition varies and on calibration to the reference sites.
Diatoms are a group of algae that have silica based cell walls. When these organisms
die, these cell walls (valves) are deposited on the substrate. Through time, subsequent
sedimentation preserves these valves. Because the valves are unique to particular
species, diatom valves present in sediment cores can be used to identify the diatom
assemblage at various points in the history of the lake. Diatoms are useable as a
biological condition indicator because many diatom species have well-defined optima and
tolerances for environmental variables such as pH, nutrients, water salinity or color
(Stoermer and Smol 1999). For this assessment, information on the environmental
conditions that favor particular diatom species can be coupled with diatom assemblage
data from reference lakes to develop a sediment "lake diatom condition" Index (LDC).
Diatom data from the surface sediments of each lake was compared to this index to
estimate the condition of the diatom assemblage. Further details on the development of
both the plankton community taxa loss model and the sediment LDC Index are
discussed in the NLA Technical Document (EPA 2010).
Results of these two models applied to the Region 10 lakes data are in Figure 7.
Results are shown as green (good), yellow (fair), red (poor) and stippled (no data). The
plankton community taxa loss model indicates that 62% of Region 10 inference lakes
are good, 10% fair, and 27% in the poor category. The sediment diatom LDC presents
a similar view with 66% in the good category. A smaller proportion is in the poor
category (3%).
15
-------
Plankton Community O/E
Region 10 Lakes Assessment
Sediment Diatom LDC
Region 10
IB-
Man-made
Natural
Region 10
Man-made
Natural
0 20 40 60 80 100
Percent Lakes
20 40 60 80 100
Percent Lakes
Figure 7. Region 10 lakes in good, fair, poor, no data, and not assessed condition classes for the
plankton community O/E and sediment lake diatom condition (LDC) indicators of biological condition.
When natural and man-made lakes are compared, plankton taxa loss model indicates a
relatively high proportion of man-made lakes in the poor category (47%) compared to
natural lakes (16%) (Figure 7). The sediment LDC index shows that most man-made
lakes are in the fair category (62%) while natural lakes are mostly of good condition for
this indicator (85%).
2. Water Quality and Trophic State
A. Water Quality
Chemical stressors have the potential to affect biota of lakes by altering their basic
environment so that required tolerable ranges are no longer present. Key chemical
stressors in this assessment are nutrients (nitrogen and phosphorus), turbidity, and
dissolved oxygen.
The nutrients nitrogen and phosphorus control algal production in lakes. These are
important because algae are the primary production that drives the biology of lakes.
Phosphorus is often the 'limiting nutrient' in lakes of the Pacific Northwest as it controls
the pace at which algae are produced. When phosphorus is exhausted by the growing
algae, the nutrient will be essentially gone from the lake and further algal growth will be
16
-------
Region 10 Lakes Assessment
limited. Likewise, small increases in phosphorus can cause rapid algal growth in lakes
that are limited by phosphorus. Excess nutrients in lakes can have negative effects on
lake biological communities, recreation, and aesthetics.
Turbidity is a measure of light scatter that is often described as murkiness or lack of
clarity. Suspended sediment or high concentrations of algae cause turbidity in lakes.
Although turbidity is a natural characteristic of lakes, human activities can decrease
water clarity by increasing sediment or nutrient levels. Nitrogen, phosphorus and
turbidity condition classes are based on regionally determined condition class
thresholds that were generated from an evaluation of reference site conditions. Details
of cut-offs and how these were calculated are available in the Technical Document
(EPA 2010). Summary statistics of relevant water quality and trophic state metrics are
in Table 4.
Table 4. Water quality summary statistics for Region 10 lakes' inference population (~1700 lakes).
Metric
Total Phosphorus
Total Nitrogen
Turbidity
Chlorophyll-a
Secchi depth*
Dissolved Oxygen**
units
ug/L
mg/L
NTU
ug/L
M
mg/L
Inference
N
1694
1694
1694
1694
1604
1694
Mean
62.22
0.65
12.18
12.37
3.89
7.72
Median
9.00
0.40
0.73
1.91
4.03
8.04
Minimum
1.00
0.01
0.24
0.07
0.04
1.00
Maximum
2670.00
7.68
574.00
194.40
36.71
11.72
Std.Dev.
186.03
0.96
58.70
25.82
2.65
1.55
* Lakes that were clear to bottom excluded from Secchi depth summary statistics.
** Dissolved oxygen value is mean of multiple measurements in the euphotic zone.
About half of the inference lakes are in good condition for nutrients, 57% of the lakes for
phosphorus and 41% of the lakes for nitrogen (Figure 8). A substantial portion of the
lakes are in the poor category for these nutrients (39% and 49%). Most man-made
lakes were in the poor category for both phosphorus and nitrogen.
17
-------
Region 10 Lakes Assessment
Total Phosphorous
RegionIO
Man-made
Natural
0 20 40 60 80 100
Percent Lakes
RegionIO
Man-made
Natural
Total Nitrogen
0 20 40 60 80 100
Percent Lakes
Figure 8. Lakes in good, fair, poor condition classes for total phosphorus and nitrogen.
Turbidity condition results are similar to those for phosphorus (Figure 9). Most of the
Region 10 inference lakes are in good condition for turbidity (58%). The man-made
lakes have a much higher proportion of lakes in the poor category compared to the
natural lakes.
18
-------
Region 10 Lakes Assessment
Turbidity
Region 10
Man-made
Natural
0 20 40 60 80 100
Percent Lakes
Figure 9. Lakes in good, fair, poor condition classes for turbidity.
Dissolved oxygen is a direct indicator of a lake's ability to support aquatic life and
aquatic organisms have specific DO requirements. In general, levels below 3 mg/L are
considered low and those below 1mg/L do not support aquatic life. For this survey, the
following EPA-recommended thresholds for DO concentration were applied: good > 5
mg/L, fair >3 to <5 and poor <3. Region-wide, dissolved oxygen measured in the
euphotic zone was high with a mean of 7.7 mg/L (Table 4). Almost all lakes are in the
good category (97%). High dissolved oxygen in the euphotic zone is an expected
result. Depths below the euphotic zone (e.g. the hypolimnion) are where low DO
concentrations are more likely to first occur. Low DO in the hypolimnion is a natural
process and is not unexpected.
B. Trophic State
The trophic state of a lake is the description of its biological productivity. Chlorophyll-a
is a measure of primary productivity of the lake and is therefore the most straightforward
indicator of lake biological productivity. Levels of nutrients and Secchi transparency
depth are indirect indicators that also give insight into the trophic state of a lake. As
described above, nutrients are often correlated to algal production as their levels can
range from very low, resulting in limiting algal production, or high, resulting in high algal
production. Secchi transparency is another measure of lake clarity. Very clear water
can indicate low levels of algal productivity while low transparency can indicate high
19
-------
Region 10 Lakes Assessment
algal presence. Thresholds of phosphorus, nitrogen, chlorophyll-a, and Secchi
transparency are used to characterize trophic condition. The thresholds used in this
assessment were selected based on input from states and are similar to those
determined by Carlson's Trophic State Index (Carlson 1977 and Carlson 1983). Table
5 shows the breakdown of parameter thresholds related to each of the four levels of
productivity.
Table 5. Carlson Trophic State Index (TSI) parameter thresholds.
Trophic State
Oligotrophic
Mesotrophic
Eutrophic
Hypereutrophic
Chlorophyll-a
(M9/L)
<2
2-<7
7-<30
>30
Total Phosphorus
(M9/L)
<10
10-<25
25-<100
>100
Total Nitrogen
(mg/L)
<0.35
0.35-<0.75
0.75 -<1. 4
>1.4
Secchi
Transparency (M)
>4
4->2
2-0.7
<0.7
Eutrophication, the process of moving to a more biologically productive state, is a
natural progression in the life of most natural lakes. The trophic conditions of lakes
range from highly productive (hypereutrophic) to very low productivity (oligotrophic). In
the Region 10 inference lake population, the chlorophyll-a measures indicate that most
of the lakes are oligotrophic (62%) followed by a relatively even distribution of the more
productive categories (Figure 10). The other three TSI thresholds for nutrients and
Secchi transparency depth also indicate dominance of oligotrophic lakes in Region 10.
ChlorA
Total P
Total N
Secchi
Hi!
I
• Oligotrophic
ElMesotrophic
D Eutrophic
DHypereutrophic
Dmissing
0%
20% 40% 60% 80% 100%
Figure 10. Percentage of inference lakes in each of four trophic categories based on
TSI thresholds for four trophic state indicators.
Comparing TSI thresholds for chlorophyll-a between the natural and the man-made
lakes, we see a large difference in the proportion of oligotrophic lakes (Figure 11). The
20
-------
Region 10 Lakes Assessment
natural lakes tend to be oligotrophic while the man-made lakes are more productive and
have higher levels of eutrophication.
Trophic Status: Chlorophyll a
RegionIO
Man-made
• Oligotrophic
QMesotrophic
n Eutrophic
D Hypereutrophic
n missing
Natural
0 20 40 60 80 100
Percent Lakes
Figure 11. Percent of lakes in four trophic classes based on TSI thresholds for chlorophyll-a.
Generally the poor classification of the lakes by trophic state is related to high nutrient
levels resulting in increased primary production. However, this relationship is more
complex as 'high' nutrients may not necessarily be responsible for a negative biological
response. Appendix 3 examines the factors that may be affecting lake production.
The possible biological response to the observed nutrient levels within Region 10 lakes
is evaluated.
3. Physical Habitat Stressors
Lakeshore and littoral habitat are important characteristics of lake ecological condition.
These habitats support the biological community by supplying food, refuge from
predators, and conditions suitable for reproduction and rearing. Shore areas also
influence nutrient cycling, production and sedimentation rates. The interface between
the lakes and human disturbance often occurs at the lakeshore where human activities
can adversely affect the lake by reducing habitat complexity. Physical habitat indicators
21
-------
Region 10 Lakes Assessment
are useful for diagnosing likely causes of ecological degradation in lakes and can be
used as benchmarks to compare future habitat changes that may result from
anthropogenic activities.
Field data collected in the terrestrial, shoreline, and littoral zones results in hundreds of
individual measurements and observations describing an array of characteristics
including bank morphology, substrate, fish concealment features, aquatic macrophytes,
terrestrial vegetative structure, and human land use and disturbance. EPA researchers
evaluated and summarized these data into four integrative physical habitat indicators of
lake condition. These four stressor indicators are described as follows:
• Littoral (shallow water zone)—metric that combines cover and structure features
used by biotic assemblages in the littoral zone. These include large woody
debris, snags, brush, overhanging vegetation, aquatic macrophytes, boulders,
and ledges.
• Lakeshore cover and structure—metric that combines structure and cover
characteristics of three layers of vegetation (tree canopy, understory, and ground
cover) present in the lakeshore zone (terrestrial and shoreline plot)
• Physical habitat complexity-metric that combines data from the littoral zone
cover estimates and the vegetation structure of the lakeshore plot.
• Human disturbance- distribution, extent, of human land use activities in the
lakeshore area and the proximity of these activities.
Condition ratings of good, fair, poor were established relative to the reference conditions
that exist within the region. Details on how these metrics were calculated and how the
condition classes were determined and assigned are described in the NLA Technical
Document (EPA 2010).
Across Region 10, 35% of the inference lake population had good lakeshore habitat
where lakeshore vegetation is complex and intact as expected by the comparisons to
the reference condition (Figure 12). An equal proportion of the lakes were in the poor
category, indicating loss of lakeshore complexity. Man-made lakes had a very high
occurrence of poor lakeshore condition as indicated by this metric. The condition of
shallow water areas was substantially better (Figure 12). Overall, the quality of the
littoral zone is good, indicating there is usually adequate quantity of shallow habitat and
that this habitat has adequate complexity to support biological assemblages. Over 68%
of the inference lakes have shallow water habitat that has structure and complexity
consistent with the reference condition.
22
-------
Region 10 Lakes Assessment
Lakeshore Habitat
Shallow Water Habitat
Region_10
Man-made
Natural
Region_10
Man-made
Natural
0 20 40 60 80 100
Percent Lakes
0 20 40 60 80 100
Percent Lakes
Figure 12. Lakes in good, fair, poor condition classes for lakeshore habitat and shallow water habitat.
The physical habitat complexity indicator is the arithmetic mean of the lakeshore and
shallow water habitat metrics and provides a more generalized view of habitat
conditions (Figure 13). About 50% of the inference lakes have good habitat vegetative
cover and complexity with good structure and cover in the littoral zone. Over 50% of the
man-made lakes are poor for overall physical habitat. The signature of human
disturbance results in fair-to poor-rating in 80% of the man-made inference lakes
(Figure 13). Less than 2% of the man-made lakes were in the good condition class for
disturbance in the lakeshore area.
23
-------
Region 10 Lakes Assessment
Physical Habitat Complexity
Lakeshore Disturbance
Region_10
Man-made
Natural
Region_10
Man-made
Natural
0 20 40 60 80 100
Percent Lakes
0 20 40 60 80 100
Percent Lakes
Figure 13. Lakes in good, fair, poor condition classes for habitat complexity and lakeshore disturbance.
4. Suitability for Recreation
Recreation indicators address the ability of lakes to support human activities such as
boating, swimming, and fishing, which are protected uses under the federal Clean
Water Act. Many factors affect the recreational quality of lakes, from aesthetic
characteristics such as water clarity and quantity of macrophytes to the quality of the
fish community that supports sport fishing. Safety for recreation is increasingly of
concern to the public. Algal toxins, pathogenic microbial organisms, and contaminants
in fish tissue are the three areas of concern for health hazards to people, pets, and
wildlife. Of these, the National Lakes Survey focuses primarily on algal toxins as
indicators of recreation suitability. The bacteria of the genus Enterococcus were
measured as an indicator of the potential for pathogens. .
A. Algal toxins
Cyanobacteria (i.e. blue-green algae) are naturally occurring algae that can produce
algal toxins. Cases of wildlife fatalities, off-flavor water and fish, and human skin rashes
have been attributed to blue-green algae exposure. Eutrophic conditions can result in
periodic cyanobacteria blooms that appear on the lake's surface as unsightly layers of
odiferous scum. Although cyanobacteria may be present, the actual production of algal
24
-------
Region 10 Lakes Assessment
toxins cannot be implied as not all cyanobacteria produce algal toxins. Determining the
presence/hazard of algal toxins is complicated by the fact that their presence is both
spatially and temporally erratic. Algal blooms are not uniform on the lake surface, but
tend to concentrate on the windward shore. Similarly, physical and chemical conditions
that promote algal blooms are ephemeral. Determining presence/hazard of algal toxins
in a one-day data collection event is therefore challenging. Three indicators were used
to assess algal toxin risk: microcystin presence - a common algal toxin, cyanobacteria
cell counts - bacteria that can produce algal toxins, and chlorophyll-a concentration - a
general measure of algal presence. Note that the last two indicators are not direct
measures of algal toxins but are surrogates for their presence. The World Health
Organization established risk thresholds that we use to assess risk to recreation
suitability (Table 6).
Table 6. World Health Organization's recreation indicator thresholds of risk associated with potential
exposure to algal toxins.
Indicator
Microcystin (M9/L)
Chlorophyll-a (M9/L)
Cyanobacteria (#/L)
Low
<10
<10
< 20,000
Moderate
10-<20
10-<50
20,000- <1 00,000
High
>20
>50
> 100,000
Microcystins, along with anatoxin-a and saxitoxins are among the most common
cyanotoxins (Graham et al. 2010). Microcystin is a liver toxin that is known to cause
tumors and is likely carcinogenic to humans. The presence of microcystin was low in
the inference lake population and only 12% of lakes had levels of microcystin above the
detection limit of 0.1 ug/L. All lakes of the inference population (100%) were rated as
having low risk to recreation suitability due to microcystin presence. The other two
measures of cyanotoxin exposure risk indicate that the inference lakes have an overall
low level of risk (Figure 14). Region-wide, the proxy indicators show over 80% of
inference lakes having low risk. For the chlorophyll-a indicator, a higher proportion
man-made lakes had high risk (30%) than for natural lakes (<2%).
25
-------
Cyanobacteria Risk
Region 10 Lakes Assessment
Chlorophyll-a Risk
RegionIO
Man-made
Natural
RegionIO
Man-made
Natural
20 40 60 80 100
Percent Lakes
20 40 60 80
Percent Lakes
100
Figure 14. Lakes within low, medium and high risk categories for algal toxin exposure based on
cyanobacteria cell counts and chlorophyll-a concentration.
B. Pathogens
Enterococci are bacteria found in soil, vegetation, and surface water that have been
contaminated by animal excrement. Most species of enterococci are not harmful to
humans, however their presence can indicate disease causing agents carried by fecal
material. Currently there are no water quality criteria or recommended thresholds for
evaluating the enterococci data collected as part of the survey. Thus, the results are
limited in their value as an indicator of the condition of lakes. For this survey, a simple
ranking of the enterococci data from all lakes of the survey was developed by EPA.
Four category rankings were determined by reviewing the data for clustering and
professional judgment. This ranking provides a relative indication of the presence of
bacteria in lakes (Table 7).
Table 7. Results of ranking Region 10 data by Enterococcus concentration categories. Samples
collected from 88 sites (inference population =1694).
Ranking (CCE/100mL)*
1 <500
2 500-1,000
3 1,000-5,000
4 >5,000
% R10 inference sites
96
1
2
1
R10 results range
Non detects- 497
515-786
1 ,698-2486
6127
*Data are expressed as Calibrator Cell Equivalents (CCE) per volume.
26
-------
Region 10 Lakes Assessment
In Region 10, most lakes are in the lowest ranking category of enterococci presence
and many of the sample lakes were below the detection limit. All of the reference lakes
sampled in Region 10 were also in the lowest category. These results add to the
consistent picture that condition of the Region's lakes is good for recreation suitability.
C. Contaminants in Fish Tissue
Fish tissue used to examine contaminant exposure was not collected as part of this
assessment. However, in the early 2000s, EPA Office of Science and Technology
conducted a National Lake Fish Tissue Study which evaluated the condition of the
nation's lakes for this important indicator of recreational suitability. This study used
similar survey design and results were applicable to the entire contiguous United States.
The Region 10 portion of this study is presented in Appendix 4. Mercury, DDTs, and
PCBs results are reported.
Summary of Findinqs: rankinq stressors
The relative extent calculation shows the stressors that have the greatest effect on the
target population. A stressor with high relative extent is both widespread and common
among lakes while stressors with low relative extent occur over a small area or
infrequently over a wide area. The relative extent for this survey is simply a ranking of
the 'poor' category results for each stressor indicator, ordered according to their
magnitude by percent of lakes. Looking across all of the stressor indicators, those
associated with water quality and trophic state had slightly higher impact on lakes of the
region followed by physical habitat indicators (Figure 15). The following are overall
summary statements of indicator results.
• Evaluation of the biological condition of Region 10 lakes relied on the combined
zooplankton/phytoplankton 0/E scores. The results showed that 62% are good,
10% fair, and 27% poor condition.
• Chlorophyll-a concentrations indicate that 12% of the lakes are hypereutrophic
with the rest having progressively lower states of eutrophication.
• Nutrients were the most extensive stressors in the region with 49% of the lakes
classified as being in poor condition for nitrogen and 39% for phosphorus.
• Three habitat stressors related to lakeshore condition (lakeshore habitat,
lakeshore disturbance and habitat complexity) were similar in their extent,
ranging from 30% to 35% in poor condition.
• Three indicators of recreational suitability support that there is a low risk of algal
toxin exposure in lakes of Region 10.
27
-------
Region 10 Lakes Assessment
Extent of Stressors- all
Total Nitrogen
Total Phosphorus
Lakeshore Habitat
Habitat Complexity
Shore Disturbance
Turbidity
Littoral Habitat
Dissolved Oxygen
49.2
0 20 40 60 80 100
Percent Lakes
Figure 15. Relative extent of poor stressor condition across all stressor metrics for all Region 10 lakes.
As shown throughout this report, the man-made lakes have generally poorer condition
than the natural lakes (Figure 16). Both habitat and water quality indicators were
substantial stressors for man-made lakes. More than 50% the man-made lakes were
classified as being in poor condition for all but two of the indicators. Also, only 45% of
man-made lakes were in the good category for biological condition compared to 72% of
the natural lakes. Natural lakes were predominantly in the good category except for the
total nitrogen indicator (40% in poor category).
28
-------
Extent of Stressors-natural
Region 10 Lakes Assessment
Extent of Stressors-manmade
Total Nitrogen
Habitat Complexity
Total Phosphorus
Lakeshore Habitat
Turbidity
Shore Disturbance
Dissolved Oxygen
Littoral Habitat
24.3
18.7
40.3
11
1.9
1.1
Lakeshore Habitat
Shore Disturbance
Total Phosphorus
Turbidity
Total Nitrogen
Habitat Complexity
Littoral Habitat
Dissolved Oxygen
20 40 60 80 10G
Percent Lakes
20 40 60 80 100
Percent Lakes
Figure 16. Relative extent of poor stressor condition across all stressor metrics for natural lakes and man-
made lake categories.
Recommendations
Several recommendations for improving the survey came to light during the course of
this analysis and from discussions with state collaborators:
• Improve site selection design to insure more reasonable error bounds in the
western ecoregions.
• Develop CIS landscape condition indicators. Our analysis of landscape metrics
in Appendix 2 shows several metrics that have good potential as indicators of
lake watershed condition. Further development of landscape indicators using
larger datasets would provide another dimension to the overall description of lake
condition.
• Add fish tissue contaminants as a recreation suitability indicator. Although some
states have monitoring programs for the tracking of toxins in fish tissue, data in
the Region 10 states is limited. Having data collected as part of this survey
would be a significant contribution to understanding this important aspect of
human health in Northwest lakes. Information on metals (mercury, arsenic,
selenium), PBDEs, and legacy pesticides, are specifically needed.
29
-------
Region 10 Lakes Assessment
• Improve ability to assess recreation suitability by sampling areas that are more
likely used for recreation. Sampling for this indicator would be more relevant at
sites adjacent to the shore rather than the center of the lake. Also, improve
ability to interpret data by collecting pathogen data that can be used to classify
the condition. Many states use E. coli. If problems with holding times can be
overcome, this would be a more useful metric for pathogens.
• Sample an adequate number of sites to allow determination of relative risk.
Although methods are available for determining relative risk, the low number of
sites in each cell of the contingency table that are used to compute the ratings
prohibits this calculation for Region 10. More sample sites would thus yield a
more meaningful analysis.
• Include a more comprehensive analysis of the diatom core in order to provide
good information on past conditions. This information would be a useful long-
term indicator.
• Research a more effective way to include invasive species assessment into the
survey. These data were included as qualitative observational information that is
not rigorously analyzed. This is a very important aspect of lake condition.
30
-------
Region 10 Lakes Assessment
References
Bailey, R. C., R. H. Morris, et al. 2004. Bioassessment of freshwater ecosystems: Using
the reference condition approach. Kluwer Academic Publishers. New York.
Carlson, R.E. 1977. A trophic state index for lakes. Limnology and Oceanography.
22:361-369.
Carlson, R.E. 1983. Discussion: Using differences among Carlson's trophic state index
values in regional water quality assessment. Water Resources Bulletin. 19:307-308.
Ebert, D., T. Wade, J. Harrison, and D. Yankee. 2000. Analytical tools interface for
landscape assessments (ATtlLA) User Guide. Version 1.004. Office of Research and
Development, U.S. Environmental Protection Agency. Las Vegas, NV.
Fraser, R.H. 1999. Sedmod: A CIS-based delivery model for diffuse source pollutants
(nonpoint source pollution). Ph.D. Dissertation. Yale University, New Haven, CT.
Graham, J. L., K. A. Loftin, et al. 2010. Cyanotoxin mixtures and taste-and-odor
compounds in cyanobacterial blooms from midwestern states. Environmental Science
and Technology. Accepted August 2010.
Hayslip, G., L. Edmond, V. Patridge, W. Nelson, H. Lee , F. Cole, J. Lamberson, and L.
Caton. 2006. Ecological condition of the estuaries of Oregon and Washington. EPA
910-R-06-001. U.S. Environmental Protection Agency, Region 10, Seattle, Washington.
Herger, L.G., G.A. Hayslip, and P.T. Leinenbach. 2007. Ecological Condition of
Wadeable Streams of the Interior Columbia River Basin. EPA-910-R-07-005. U.S.
Environmental Protection Agency, Region 10, Seattle, Washington.
Hughes, R. M. 1995. Defining acceptable biological status by comparing with reference
conditions. Pages 31-47 in Biological Assessment and Criteria: Tools for water
resource planning and decision making. W.S. Davis and T. P. Simon. Lewis Publishers.
London.
Omernik, J. M. 1987. Ecoregions of the conterminous United States. Annals of the
Association of American Geographers 77: 118-125.
Overton, W. S., D. White, et al. 1990. Design report for EMAP, Environmental
Monitoring and Assessment Program. EPA/600/3-91/053. U.S. Environmental
Protection Agency. Washington, D.C.
Renard, K.G., G.R. Foster, G.A. Weesies, O.K. McCool, and D.C. Yoder. 1997.
PrediSoil Erosion by Water: A Guide to Conservation Planning with the Revised
Universal Soil Loss Equation (RUSLE). Agriculture Handbook No. 703. U.S. Department
of Agriculture, Agricultural Research Service, Washington, D.C. 404 pp.
31
-------
Region 10 Lakes Assessment
Rowan, J.S., R.W. Duck, O.M. Bragg, A.R. Black, and I.S.M.E.J. Cutler, and P.J. Boon.
(2006). Development of a technique for Lake Habitat Survey (LHS) with applications for
the European Union Water Framework Directive. Aquatic Conserv: Mar. Freshw.
Ecosyst. 16: 637-657 (2006) 16: 637-657.
Stevens, D.L., Jr. and A.R. Olsen 2004. Spatially balanced sampling of natural
resources. Journal of the American Statistical Association 99:262-278.
Stoddard, J. L, D. P. Larsen, et al. 2006. Setting expectations for the ecological
condition of streams: the concept of reference condition. Ecological Applications 16(4):
1267-1276.
Stoermer E.F. and J.P. Smol. 1999. Applications and uses of diatoms: prologue. In
Stoermer E.F. and Smol J.P. (eds) The diatoms: Applications for the
Environmental and Earth Sciences, Cambridge, Cambridge University Press, pp 38.
U.S.E.P.A. and U.S.G.S. 2009. NHDPIus users guide. (Tommy Dewald, Project
Manager, EPA Office of Water), http://www.horizon-systems.com/nhdplus/.
USEPA. 2003. Level III ecoregions of the continental United States (revision of
Omernik, 1987). U.S. EPA National Health and Environmental Effects Research
Laboratory. Corvallis, Oregon. Map M-1, various scales.
USEPA. 2007. Survey of the Nation's Lakes. Field Operations Manual. EPA 841-B-07-
004. U.S. Environmental Protection Agency. Washington, DC.
USEPA. 2009a. 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, DC. April 2010.
USEPA. 2009b. Survey of the Nation's Lakes: Integrated Quality Assurance Project
Plan. EPA/841-B-07-003. U.S. Environmental Protection Agency, Office of Water and
Office of Research and Development. Washington, DC.
USEPA. 2010. National Lakes Assessment: Technical Appendix Data Analysis
Approach. EPA 841-R009-001a. U.S. Environmental Protection Agency Office of Water
and Office of Research and Development. Washington, D.C.
Van Remortel, R.D., R.W. Maichle, D.T. Heggem and A.M. Pitchford. 2005. Automated
CIS watershed analysis tools for RUSLE/SEDMOD soil erosion and sedimentation
modeling. EPA/600/X-05/007. USEPA, Washington, DC.
32
-------
Region 10 Lakes Assessment
Appendix
Appendix 1 . List
SITEJD
NLA06608-0005
NLA06608-0085
NLA06608-0241
NLA06608-0277
NLA06608-0497
NLA06608-0561
NLA06608-0581
NLA06608-0597
NLA06608-0650
NLA06608-0769
NLA06608-0837
NLA06608-0961
NLA06608-1281
NLA06608-1329
NLA06608-1473
NLA06608-1521
NLA06608-1537
NLA06608-1793
NLA06608-1930
NLA06608-1985
NLA06608-1989
NLA06608-2005
NLA06608-2305
NLA06608-2497
NLA06608-2801
NLA06608-2954
NLA06608-3121
NLA06608-3157
NLA06608-3313
NLA06608-3329
NLA06608-0049
NLA06608-0290
NLA06608-0306
NLA06608-0402
NLA06608-0406
NLA06608-0614
NLA06608-0625
NLA06608-0658
NLA06608-0677
1.
Sampled
Lakes in
Region
10
of sampled probability lakes including design coordinates
State Weight
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
ID
OR
OR
OR
OR
OR
OR
OR
OR
OR
15.010328
9.195787
3.738934
3.738934
9.195787
2.425646
2.425646
2.425646
4.030970
4.030970
3.395574
3.660791
233.572732
4.030970
2.425646
3.395574
15.010328
4.030970
3.738934
2.425646
3.738934
3.395574
9.914039
3.660791
3.395574
13.922860
4.030970
4.030970
3.395574
3.660791
16.398147
3.999259
4.084627
168.547626
4.403663
3.999259
168.547626
3.999259
3.709520
Longitude
(dd)
-114.840386
-116.141432
-111.729262
-113.400612
-111.815938
-116.089322
-114.382789
-116.665495
-111.396362
-116.602150
-113.923741
-116.686557
-116.155618
-116.169740
-116.893380
-111.853579
-116.638410
-116.564348
-111.473476
-116.798693
-114.891000
-116.112474
-116.527985
-116.970121
-112.689224
-111.603733
-116.462359
-115.798714
-111.836249
-116.824496
-117.153355
-120.525686
-119.413348
-124.078093
-121.704430
-123.268618
-118.185992
-124.079610
-118.446839
Latitude
(dd)
43.929209
43.460454
42.291543
42.678174
42.107653
44.948436
43.252639
43.547058
44.642099
47.465617
43.325359
47.489395
48.135006
45.068128
47.890852
42.233175
48.161727
47.520730
44.100512
47.448688
42.202900
42.297374
48.184430
48.008220
42.085663
44.023613
44.964527
43.549425
42.123193
48.457411
45.062327
42.195652
43.417883
42.889749
45.180381
43.378645
44.954294
44.023835
42.772394
Area
(ha)
19.5
33.4
170.2
3395.5
34.2
2018.4
1443.3
3571.2
2459.2
399.6
81.3
91.9
6.2
148.1
370.3
61.5
17.4
201.8
144.2
11029.2
393.0
56.6
39.3
54.5
50.2
19.1
211.4
332.4
52.3
70.5
12.9
59.2
107.8
7.5
141.9
52.7
6.8
60.6
89.8
(90 total).
Major
ecoregion
WMT
XER
XER
XER
XER
WMT
XER
XER
WMT
WMT
XER
WMT
WMT
WMT
WMT
XER
WMT
WMT
XER
WMT
XER
XER
WMT
WMT
XER
XER
WMT
WMT
XER
WMT
WMT
WMT
XER
WMT
WMT
WMT
WMT
WMT
XER
Level III
ecoregion
16
12
80
12
13
16
12
12
17
15
12
15
15
16
15
13
15
15
12
15
80
80
15
15
13
12
11
16
13
15
11
9
80
1
4
78
11
1
80
33
-------
Region 10 Lakes Assessment
Appendix 1 continued. List of sampled probability lakes including design coordinates (90 total).
SITEJD
NLA06608-0678
NLA06608-0870
NLA06608-0881
NLA06608-0933
NLA06608-0934
NLA06608-1058
NLA06608-1073
NLA06608-1190
NLA06608-1266
NLA06608-1426
NLA06608-1445
NLA06608-1446
NLA06608-1638
NLA06608-1894
NLA06608-1958
NLA06608-2082
NLA06608-2438
NLA06608-2450
NLA06608-2481
NLA06608-2673
NLA06608-2726
NLA06608-0033
NLA06608-0081
NLA06608-0086
NLA06608-0193
NLA06608-0209
NLA06608-0337
NLA06608-0449
NLA06608-0529
NLA06608-0593
NLA06608-0641
NLA06608-0721
NLA06608-0785
NLA06608-0849
NLA06608-1041
NLA06608-1057
NLA06608-1217
NLA06608-1297
NLA06608-1425
NLA06608-1489
NLA06608-1617
NLA06608-1873
NLA06608-2134
NLA06608-2193
State
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
Weight
4.403663
3.999259
4.403663
4.403663
1.750362
4.403663
10.830667
4.403663
15.210134
4.403663
4.084627
10.830667
3.999259
10.830667
1.750362
4.403663
10.046007
16.398147
168.547626
16.398147
10.830667
32.549373
32.549373
8.741017
7.938297
2.337088
7.938297
21.498248
2.337088
32.549373
21.498248
7.938297
7.363184
2.337088
7.363184
8.741017
8.741017
8.107748
32.549373
21.498248
7.938297
8.741017
21.498248
32.549373
Longitude
(dd)
-121.780483
-122.046378
-118.045641
-118.150427
-122.038246
-122.214210
-117.272370
-123.300076
-119.997611
-124.246005
-118.852169
-122.017536
-121.873209
-121.745625
-122.421606
-122.600822
-119.563401
-124.179044
-119.392762
-118.685037
-121.766543
-121.299060
-122.124168
-122.426035
-121.175840
-122.083679
-122.656243
-117.691524
-119.363985
-122.327709
-117.664050
-122.405638
-119.570872
-124.632527
-119.592605
-121.094003
-117.332380
-119.648323
-122.972559
-122.569031
-122.705033
-123.264720
-122.743569
-124.035496
Latitude
(dd)
43.963587
44.316686
44.680299
43.927628
43.736127
42.364916
45.229364
44.087941
42.119982
43.452205
42.918354
43.796302
44.371781
44.026655
43.662976
42.151396
42.066805
43.631644
43.988628
44.306582
43.713787
47.311084
48.226120
45.616571
48.704386
47.576010
48.393824
48.135804
47.184414
48.708678
48.418773
46.497287
46.871389
48.090401
47.728697
47.266902
48.273757
46.693467
47.192047
46.986303
47.571772
47.487995
45.892663
46.419506
Area
(ha)
102.2
63.7
911.8
716.8
2443.7
477.1
24.3
3230.3
13.8
132.3
273.1
27.7
90.7
32.9
1062.1
256.5
27.1
14.5
5.9
12.7
40.9
15.2
10.0
101.3
81.4
1934.3
51.9
48.2
2575.6
10.3
27.0
84.0
50.9
3036.2
75.2
1816.3
194.6
219.3
14.5
22.2
93.2
1551.9
31.8
17.6
Major
ecoregion
WMT
WMT
WMT
WMT
WMT
WMT
WMT
WMT
XER
WMT
XER
WMT
WMT
WMT
WMT
WMT
XER
WMT
WMT
WMT
WMT
WMT
WMT
WMT
WMT
WMT
WMT
WMT
XER
WMT
WMT
WMT
XER
WMT
XER
WMT
WMT
XER
WMT
WMT
WMT
WMT
WMT
WMT
Level III
ecoregion
4
4
11
11
4
4
11
3
80
1
80
4
4
4
4
78
80
1
11
11
9
77
2
3
77
2
2
15
10
2
15
4
10
1
10
77
15
10
2
2
2
1
3
1
34
-------
Region 10 Lakes Assessment
Appendix 1 continued. List of sampled probability lakes including design coordinates (90 total).
SITEJD
NLA06608-2241
NLA06608-2257
NLA06608-2513
NLA06608-2753
NLA06608-2833
NLA06608-3153
NLA06608-3265
State
WA
WA
WA
WA
WA
WA
WA
Weight
8.741017
225.045285
2.337088
2.337088
7.363184
7.938297
21.498248
Longitude
(dd)
-117.318999
-122.784397
-122.561921
-117.688988
-120.164879
-122.816780
-117.408610
Latitude
(dd)
48.154021
47.409308
47.132856
47.570687
47.918804
48.660289
48.056480
Area
(ha)
120.4
7.2
441.5
49.6
76.7
73.9
28.2
Major Level III
ecoregion ecoregion
WMT
WMT
WMT
XER
XER
WMT
WMT
15
2
2
10
10
2
15
35
-------
Region 10 Lakes Assessment
Appendix 2. Analysis of Landscape Metrics
Introduction
CIS was used to generate landscape indicators that integrate conditions over the
broader watershed contributing to lake condition. Landscape metrics were generated
uniformly for all sample sites, which is useful when lake size prohibits adequate field
collection of physical habitat data from lake shoreline zones. Our goal is to identify and
test landscape metrics as indicators of lake watershed condition.
Objectives
1) Evaluate a large group of available landscape metrics to identify a shorter list of
those with the most potential for estimating lake watershed condition.
2) Using the candidate indicators from the short list that perform best, describe lake
watershed condition for the Western Mountains portion of Region 10 NLA sites.
Landscape metrics calculated at the scale of the entire basin may not be closely tied to
nearshore habitat condition, which has been found to have a greater influence on lake
condition. We calculated landscape metrics at various buffer distances from the lake
shore to find the most useful scale for each metric.
Physical habitat indicators used in the NLA will be used to evaluate the landscape
indicators. Generally, we expect a relationship between physical habitat indicators and
landscape indicators. Lakes within watersheds with extensive watershed-scale
disturbance are likely to have higher lakeshore disturbance as well as reduced lake
condition. Conversely, lakes within watersheds with low levels of human disturbance
are likely to have lower levels of human disturbance in proximity to the lakeshore with
natural/intact lakeshore vegetation and littoral cover complexity that result in higher lake
condition.
Identifying Best Metrics
Methods
This analysis uses watersheds from the 101 lakes sampled for the Region 10 portion of
the NLA (90 probability and 11 handpicked sites). The initial list of landscape metrics
included about 60 basin-scale variables within the categories shown on Table 3. The
landcover type metrics were also calculated at three buffer widths (5km, 2km, and
200m) for each sample lake. Landscape metrics with the best potential to be used as
indicators of lake watershed condition have the following characteristics:
1) an obvious relation to human disturbance (e.g. land use v. geomorphic metrics)
36
-------
Region 10 Lakes Assessment
2) a logical relation to lake disturbance/response indicators
3) A reasonable range of values are represented in the data set. For example
metrics that have many zero values such as %wetlands are not useful.
4) Limited autocorrelation so that each metric included is informative and not
redundant.
Based on these desired characteristics the following steps were taken to winnow the list
of landscape metrics:
-Check strength of association between landscape metrics generated at the basin-wide
scale with lake indicators (physical habitat, water quality, biology response) using non-
parametric Spearman rank correlation. Also, plot and review scatter diagrams for data
distribution. Identify candidate list of metrics based on strength and pattern of the
associations.
-Review the candidate metrics for logical relationships with lake disturbance/response
indicators.
-Review the candidate metrics for autocorrelation to avoid redundancy.
-Recalculate the candidate metrics at the three buffer distances and re-check strength
of association (non-parametric Spearman rank correlation) of each. From these results
evaluate the best buffer distance for each of the candidate metrics.
-Review the modified candidate metrics for their adequacy of describing ecological
condition and determine final list of metrics to carry forward into indicator development.
Results
Following these steps, a short list of the best landscape metrics in terms of relation to
lake disturbance/response indicators was determined (Table A1). The results of the
correlations to disturbance/response indicators are in Table A2. Other correlations
used for data exploration are at the end of this appendix.
Table A1. Final list of best landscape metrics based on analysis of 101 Region 10 watersheds from the
NLA.
Metric
Forest cover
Scrub-shrub cover
Potential Unit Grazing
RUSLE Cover Factor
Units
% total cover
% total cover
unitless
unitless
buffer width
200m
2Km
Basin-wide
Basin-wide
37
-------
Region 10 Lakes Assessment
Table A2. Spearman rank correlation rvalues between best landscape metrics and chemical, biological,
and physical habitat lakes metrics. Significant (p<.05) yet weak relationships are in bold blue <|0.5| and
moderate to high correlations in bold italic red >|0.5|.
Forest Cover
(200m)
Scrub-shrub
Cover (2km)
Potential Unit
Grazing (basin)
RUSLE Cover
Factor (basin)
*Epi.
DO
0.41
-0.44
-0.04
-0.24
Cond
-0.59
0.52
0.41
0.38
Turb
-0.75
0.54
0.36
0.57
N
total
-0.53
0.43
0.39
0.33
P
total
-0.62
0.54
0.34
0.42
Chl-a
-0.54
0.28
0.50
0.23
Secc Phyto shore
depth OE Dis.
0.70 0.43
-0.47 -0.18
-0.42 -0.28
-0.38 -0.33
-0.28
0.12
0.34
-0.01
shore
cover
0.65
-0.66
-0.01
-0.46
Lit
Cov.
0.50
-0.57
0.00
-0.30
Habitat
complex.
0.64
-0.68
0.00
-0.44
Metric abbreviations: dissolved oxygen mean of values collected in upper 2m of water column, conductivity, turbidity,
total nitrogen, total phosphorous, chlorophyll-a, Secchi depth, phytoplankton O/E, shoreline disturbance (RDIS),
lakeshore cover (LITCVR), littoral cover (LITCVR), habitat complexity (LITRIPCVR).
Two other metrics, total agriculture cover and agriculture on slopes >9%, had
reasonably good correlations but data were dominated by zero values. Likely that a
more robust data set that included more sites in the xeric cluster would show these to
be more useful. The results of the Sediment Model run for each watershed was tested
but it had an inconsistent relation to lake response /distribution metrics. The model
incorporates geomorphic metrics (slope) which effectively swamps out the effect of the
actual sediment load delivered to the lake. Thus, this metric is not useful for this
application. However, the cover component of the Model 'RUSLE Cover Factor' was
identified as a useful metric. The four retained metrics are described below.
Forest Cover:
Forest cover areas are characterized by tree cover (natural or semi-natural woody
vegetation, generally greater than 6 meters tall); tree canopy accounts for 25-100
percent of the cover. This cover class includes the following cover codes of the NLCD:
41. Deciduous Forest - areas dominated by trees where 75% or more of the tree
species shed foliage simultaneously in response to seasonal change.
42. Evergreen Forest - areas dominated by trees where 75% or more of the tree
species 'maintain their leaves all year. Canopy is never without green foliage.
43. Mixed Forest - Areas dominated by trees where neither deciduous nor evergreen
species represent more than 75% of the cover present.
This metric is calculated for the portion of the basin within 200m of the lakes shoreline
and is expressed as percent of the total cover. The forest cover metric is positively
correlated to lake physical habitat, water clarity, and biological response indicators.
Scrub-shrub Cover:
Shrublands are areas characterized by natural or semi-natural woody vegetation with
aerial stems, generally less than 6 meters tall, with individuals or clumps not touching to
interlocking. Both evergreen and deciduous species of true shrubs, young trees, and
trees or shrubs that are small or stunted because of environmental conditions are
38
-------
Region 10 Lakes Assessment
included. This cover class includes code 51 of the NLCD coverage defined as: Areas
dominated by shrubs; shrub canopy accounts for 25-100% of the cover. Shrub cover is
generally greater than 25% when tree cover is less than 25%. Shrub cover may be less
than 25% in cases when the cover of other life forms (e.g. herbaceous or tree) is less
than 25% and shrubs cover exceeds the cover of the other life forms.
The scrub-shrub metric is calculated for the 2km buffer area adjacent to the lakeshore
and is expressed as percent of the total cover. As expected, this metric has a negative
correlation to lake physical habitat, water clarity, and biological response indicators.
Potential Unit Grazing:
The Potential Unit Grazing (PUG) is an indicator of the intensity of potential cattle/calf
grazing. This metric is based on estimations of cattle usage coupled with estimations of
'cow habitat' and requires calculation of several grids. The following is a brief
description of how this metric is calculated.
1) A Cow Density grid is generated using USDA agriculture census data reported by
counties on the total number of grazing cow-calves.
2) A Potential Cow Habitat grid is generated from methods developed by EPA
Region 10 (Pers. Comm. Peter Leinenbach) using five inputs: land ownership,
land cover, proximity to water, topographic position grid index, and slope.
3) A relative risk weighting factor is developed for the Potential Cow Habitat grid to
define the relative intensity of the habitat weights in each habitat grid cell.
Applying these weights to the grid results in a new 'Potential Cow Habitat Usage'
grid.
4) The final metric Potential Unit Grazing (PUG) grid calculated as the Cow Density
Grid multiplied by the Potential Cow Habitat Usage grid.
This metric is calculated for the entire basin and is unitless. The Potential Unit Grazing
indicator has a negative correlation to lake physical habitat, water clarity, and biological
response indicators. In general, areas with the highest potential for water quality
impacts due to grazing are flat non-protected grasslands, within 90 meters of a water
source, within counties with high cattle densities and low available potential cow habitat.
RUSLE Cover Factor.
The RUSLE (Revised Universal Soil Loss Equation) model predicts potential surface
soil erosion across the landscape. One input to this model is the RUSLE Cover Factor
(i.e., RUSLE C), which is used to reflect the effect of agricultural cropping and landcover
management practices on erosion rates. Specifically, RUSLE C represents the effects
of plants, soil cover, soil biomass, and soil disturbing activities on erosion. Although no
specific reference is available to obtain these values for landscape scale modeling,
various references can provide estimates established at the plot scale (listed at the end
of this Appendix). USING the SEDMOD model, RUSLE C plot scale measurements
obtained from literature were applied across the landscape and estimates of weighted
average watershed RUSLE C conditions were calculated.
39
-------
Region 10 Lakes Assessment
The RUSLE Cover Factor is unitless and is calculated at the basin scale. It is
negatively correlated to lake physical habitat, water clarity, and biological response
indicators in the data set of Region 10.
Indicator Development
The four landscape metrics identified as having the most substantial relation to lake
watershed condition were carried forward into the indicator development phase.
Methods
Data for the indicator development analysis were restricted to sites within the Western
Mountains cluster. This was necessary as there were insufficient numbers of reference
sites from the Xeric cluster in Region 10 to include sites from this portion of the Region.
A total of 17 reference sites (10 handpicked and 7 from probability site dataset) were
available for determining condition thresholds for the Western Mountains cluster in
Region 10. Recall, all probability and hand-picked sites in the NLA are evaluated for
chemical and lakeshore characteristics and qualified as reference or non-reference.
Because we are developing thresholds with only the Western Mountains reference sites
the final analysis of the condition was only applied to the probability sites from this same
cluster. Of the 90 probability sites in Region 10, 65 sites are within the Western
Mountains cluster.
Thresholds were based on the distribution of values in the set of 17 reference sites and
whether the correlation with the indicator is positive or negative. For indicators where
high values indicate a better condition (e.g. forest cover), we used the 25th percentile of
the distribution of the reference sites values to distinguish between "least disturbed"
(similar to reference site condition) and "somewhat disturbed" (somewhat different from
set of reference condition values). The 5th percentile was used to distinguish between
"somewhat disturbed" and "disturbed" (very different from the set of reference sites).
For indicators where high values indicate disturbance or poorer condition (e.g. scrub-
shrub cover), the thresholds were reversed. The 75th percentile of the distribution of
the reference site values was used to distinguish between "least disturbed" and
"somewhat disturbed" condition, and the 95th percentile was used to distinguish
between "somewhat disturbed" and "disturbed" condition. This scoring was
conservative to account for the fact that, although minimally disturbed, reference sites
may have some level of human disturbance. The thresholds determined for each of the
four landscape metrics are in Table 3A and box plots showing the range of values in the
reference data for each metric are in FigurelA.
40
-------
Region 10 Lakes Assessment
Table 3A. Data range percentiles (calculated directly) used to define thresholds between Least,
Intermediate, and Most disturbed for four landscape metrics based on Western Mountain region reference
sites.
Indicator
Forest cover
Scrub-shrub cover
Potential Unit Grazing
*RUSLE Cover Factor
buffer extent
200m
2 Km
basin-wide
basin-wide
refsite n
17
17
17
15
Correlation
type
positive
negative
negative
negative
5%
33.36
0.00
0.00
0.02
25%
65.19
0.28
0.03
0.03
75%
93.91
17.54
0.15
0.06
95%
97.35
25.28
0.43
0.09
Two extreme values omitted before calculation of thresholds for the RUSLE Cover Factor.
41
-------
Region 10 Lakes Assessment
Forest Cover 200m buffer
Scrub-shrub Cover 2km buffer
110
100
90
E
o
70
60
50
40
30
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
-0.05
Potential Unit Grazing basin
n Median = 77.2472
0 25%-75%
= (65.2, 93.9)
X non-outl.range
(33.4, 100)
E
<5
o
u
.a
D
O
CO
26
24
22
20
18
16
14
12
10
8
6
4
2
0
-~>
D
n Median = 0.081
D 25%-75%
= (0.030, 0.148)
I Non-Outlier Rang
= (2.5E-5, 0.248)
o Outliers
0.10
0.09
0.08
« 0.07
<5
O 0.06
U
LLJ
CO 0.05
0.04
0.03
0.02
Q Median = 6.2
0 25%-75% = (0.3, 17.5)
X Non-Outlier Range
= (0, 25.3)
RUSLE Cover Factor.extremes omitted
T
D Median = 0.042
D 25%-75%
= (0.034, 0.059)
Non-Outlier Range
= (0.024, 0.093)
Figure 1A. Range of values across 17 Western Mountain region reference sites for four best landscape metrics. RUSLE C factor presented
without extreme values (1X1=15).
42
-------
Region 10 Lakes Assessment
Results
The thresholds calculated for the four landscape metrics were applied to the western
mountain Region 10 data yielding the relative percents of least disturbed, somewhat
disturbed, and most disturbed sites for the 65 sites (Figure 2A). These results were
compared to the physical habitat indicator results (Figure 3A) for the same western
mountain sites. The best performing landscape indicator was forest cover calculated at
the 200m buffer. This buffer yielded similar results for the 65 sites as did the habitat
complexity indicator (combined metric of littoral/lakeshore cover) where about 55% of
the sites are in the least disturbed category. The scrub-shrub and the RUSLE C factor
indicators show a similar pattern, with a large majority of sites classified in the least
disturbed category.
The Potential Unit Grazing results were most similar to the shoreline disturbance
indicator of the physical habitat indicators. Recall that the indicator of shoreline
disturbance is based on evidence of human disturbance in proximity to the shore. Both
of these indicators estimate a much lower portion of the sites in the least disturbed
condition category.
Future Work
The four indicators identified in this appendix yielded results similar to the physical
habitat indicators for the Western Mountain cluster of the Region 10 portion of the NLA.
It would be useful to further test these indicators with a larger dataset of reference sites
and in different ecoregions. These have the potential to be used for the upcoming lakes
survey with further refinement. They may also serve as a reasonable substitution for
field data from lakes that are too large to reasonably collect field data on physical
habitat.
43
-------
Region 10 Lakes Assessment
Forest
Scrub-shrub
Rusle Cover 13.8
Potential Grazing
H Least
• Intermediate
• Most
0 20 40 60 80 100
Figure 2A. Percent of least, intermediate, and most disturbed condition for sites of the Western Mountain
cluster of the Region 10 NLA based on landscape condition indicators (n=65).
Lakeshore Cover
Littoral Cover
Habitat Complex.
Shore Disturb.
H Least
• Intermediate
• Most
0 20 40 60 80 100
Figure 3A. Percent of least, intermediate, and most disturbed condition for sites of the Western Mountain
cluster of the Region 10 NLA based on four physical habitat condition indicators (n=64).
44
-------
Region 10 Lakes Assessment
Sources used to develop estimates for the Rusle C factor.
Dissmeyer, G. E., and G. R. Foster. 1980. A Guide for Predicting Sheet and Rill Erosion
on Forest Land. Technical Publication SA-TP 11, USDA Forest Service, Atlanta, GA.
Dissmeyer, G. E., and G. R. Foster. 1981. Estimating the cover-management factor (C)
in the universal soil loss equation for forest conditions. Journal of Soil and Water
Conservation 36:235-240.
Eslinger, David L, H. Jamieson Carter, Ed Dempsey, Margaret VanderWilt, Beverly
Wilson, and Andrew Meredith. 2005. "The Nonpoint-Source Pollution and Erosion
Comparison Tool." NOAA Coastal Services Center, Charleston, South Carolina.
Accessed [Month Year] at http://www.csc.noaa.gov/nspect/.
Wischmeier, W. H., and D. D. Smith. 1978. Predicting Rainfall and Erosion Losses: A
Guide to Conservation Planning. Agriculture Handbook No. 537, U. S. Department of
Agriculture, Washington, D.C.
Yang, D. W., S. Kanae, T. Oki, T. Koike, and K. Musiake. 2003. Global potential soil
erosion with reference to land use and climate changes. Hydrological Processes
17:2913-2928.
Zaluski, M. H., J. J. Consort, and S. B. Antonioli. 2004. Soil erosion and deposition
modeling using ArcGIS. in M. Aides, T. Hang, and L. M. Deschaine, editors. 2004
Business and Industry Symposium. The Society for Modeling and Simulation
International, San Diego, California, (http://www.scs.org/getDoc.cfm?id=1711, accessed
May 16, 2006)
45
-------
Region 10 Lakes Assessment
Spearman Correlation results for landscape and lakeshore condition metrics to lake condition metrics. Significant
correlations (p<.05) are red bold.
Total Ag. Cover (200m)
Forest Cover (200m)
Scrub-shrub Cover (2km)
Agr. on >9% slope (basin)
Sediment Model (basin)
Potential Unit Grazing (basin)
RUSLE Cover Factor (basin)
Shoreline disturbance
Lakeshore Cover Index
Littoral Cover Index
Lit. -Rip. Cover Index
EpiDO
-0.27
0.41
-0.44
-0.16
0.14
-0.04
-0.24
-0.08
0.51
0.41
0.50
Cond.
0.42
-0.59
0.52
0.59
-0.37
0.41
0.38
0.25
-0.35
-0.28
-0.34
Turb.
0.53
-0.75
0.54
0.45
-0.29
0.36
0.51
0.31
-0.59
-0.43
-0.57
N total
0.33
-0.53
0.43
0.34
-0.45
0.39
0.33
0.25
-0.41
-0.32
-0.40
P total
0.43
-0.62
0.54
0.34
-0.38
0.34
0.42
0.30
-0.48
-0.39
-0.49
Chl-a
0.47
-0.54
0.28
0.39
-0.32
0.50
0.23
0.24
-0.32
-0.19
-0.31
Secchi mean
-0.51
0.70
-0.47
-0.40
0.40
-0.42
-0.38
-0.29
0.57
0.45
0.57
Phyto.OE
-0.30
0.43
-0.18
-0.23
0.27
-0.28
-0.33
-0.23
0.21
0.23
0.24
Spearman correlation results for landscape and lakeshore condition metrics to geophysical characteristic metrics
Total Ag. Cover (200m)
Forest Cover (200m)
Scrub-shrub Cover (2km)
Agr. on >9% slope (basin)
Sediment Model (basin)
Potential Unit Grazing (basin)
RUSLE Cover Factor (basin)
Shoreline disturbance
Lakeshore cover
Littoral Cover
Habitat complexity
W.S.
Area
0.36
-0.40
0.43
0.54
-0.06
0.06
0.29
0.14
-0.32
-0.31
-0.34
Lake
Polygon
Area
0.30
-0.27
0.13
0.40
-0.01
0.09
0.20
0.14
-0.18
-0.18
-0.19
Lake
Surface
Proport.
-0.25
0.32
-0.58
-0.37
0.11
-0.01
-0.25
-0.08
0.32
0.30
0.35
Elev.
mean
0.03
-0.08
0.47
-0.04
0.07
-0.41
0.19
-0.17
-0.37
-0.33
-0.39
Slope
mean
-0.06
0.22
0.12
-0.09
0.71
-0.44
-0.19
-0.18
-0.09
-0.03
-0.10
Precip.
max.
-0.35
0.49
-0.59
-0.32
0.55
-0.16
-0.30
-0.17
0.44
0.42
0.44
Precip.
mean
-0.48
0.64
-0.74
-0.56
0.56
-0.23
-0.41
-0.23
0.55
0.49
0.55
46
-------
Region 10 Lakes Assessment
Spearman correlation results to check for autocorrelation for landscape and lakeshore metrics.
Total Ag. Cover (200m)
Forest Cover (200m)
Scrub-shrub Cover (2km)
Agr. on >9% slope (basin)
Sediment Model (basin)
Potential Unit Grazing (basin)
Rusle Cover Factor (basin)
Shoreline disturbance
Lakeshore cover
Littoral Cover
Habitat Complexity
Total
Ag.
1.00
-0.60
0.36
0.62
-0.17
0.24
0.31
0.25
-0.40
-0.38
-0.41
Forest
-0.60
1.00
-0.67
-0.62
0.35
-0.34
-0.67
-0.28
0.65
0.50
0.64
Scrub
0.36
-0.67
1.00
0.47
-0.22
0.06
0.55
0.12
-0.66
-0.57
-0.68
Ag>9
0.62
-0.62
0.47
1.00
-0.22
0.41
0.43
0.21
-0.32
-0.30
-0.33
Sed.Mod.
-0.17
0.35
-0.22
-0.22
1.00
-0.32
-0.10
-0.16
0.10
0.11
0.10
PUG
0.24
-0.34
0.06
0.41
-0.32
1.00
0.14
0.34
-0.01
0.00
0.00
RUSLE
c
0.31
-0.67
0.55
0.43
-0.10
0.14
1.00
-0.01
-0.46
-0.30
-0.44
Shore
Distrub
0.25
-0.28
0.12
0.21
-0.16
0.34
-0.01
1.00
-0.17
-0.21
-0.19
Shore
cover
-0.40
0.65
-0.66
-0.32
0.10
-0.01
-0.46
-0.17
1.00
0.66
0.96
Lit.Cov.
-0.38
0.50
-0.57
-0.30
0.11
0.00
-0.30
-0.21
0.66
1.00
0.81
Habitat
compl.
-0.41
0.64
-0.68
-0.33
0.10
0.00
-0.44
-0.19
0.96
0.81
1.00
47
-------
Region 10 Lakes Assessment
Appendix 3. Additional Nutrient Analysis for Region 10 NLA
Lakes
Recall that elevated nitrogen and phosphorus levels (categorized as "poor" in Figure 8)
were observed in many of the Region 10 lakes. Despite the "poor" classification for
some of the lakes, it is possible that these high nutrient levels are not resulting in
increased primary productivity (i.e., excessive algal growth) and therefore they might not
be resulting in a negatively impacted biological response. This appendix evaluates the
possible biological response to the observed nutrient levels within Region 10 lakes.
Background Information
The Concept of the Limiting Nutrient
Research has shown that algal growth in a lake will be limited if at least one of the
nutrients (typically either nitrogen or phosphorus) is at or below the literature values for
the Michaelis-Menten half-saturation constants. Specifically, the relationship between
the algal growth rate and concentration of a substrate (nutrients) can be described by
the following empirical model:
N
where GN is the algal growth rate depending on nutrient supplies, Gmax is algal growth
rate at optimum temperature, light and nutrient conditions, N is the substrate (nutrient)
concentration, and Ks is the nutrient concentration at which the algal growth rate is one
half (0.5) the maximum rate and is referred to as the Michaelis-Menten half-saturation
constant. As can be inferred from this equation, where in-lake nutrient concentrations
are low, algal growth is inhibited, while at high concentrations algal nutrient demands
are fully met and growth is limited by other factors (i.e., light availability and
temperature) (Thomann and Mueller 1987). While algal growth will still occur at low
concentrations, algal growth will be drastically reduced.
Algal growth will not be limited by nitrogen or phosphorus, however, if water column
nutrients are present in concentration exceeding five times the respective Michaelis-
Menten half-saturation constants (Thomann and Mueller 1987). Typical half-saturation
constants for phosphorus and nitrogen are 8 ug/l and 25 ug/l, respectively (EPA/600/3-
85/040). Accordingly, nutrient concentrations that should saturate algal growth
demands are defined as 40 ug/l and 125 ug/l, phosphorus and nitrogen, respectively.
There are several ways that the nutrient limitation term can be expressed (i.e.,
Multiplicative, Minimum, and Harmonic Mean) (Chapra 1997). The most common
approach used is where the nutrient in shortest supply is expected to control growth:
GN = min{GN-phosphorus, GN-nitrogen)
48
-------
Region 10 Lakes Assessment
Nitrogen to Phosphorus Ratios
Algae incorporate inorganic nutrients from the water in proportion to their cellular
stoichiometry (Chapra 1997). Specifically, the Redfield ratio is often used to judge the
requirements of algae:
Oxygen:Carbon:Nitrogen:Phosphorus = 109:41:7.2:1 (by weight)
As can be seen in the equation above, algal cell stoichiometry is well represented by a
Nitrogen to Phosphorus (N/P) ratio of 7. Thomann and Mueller (1987) reported that
nitrogen could cause algal grown limitations when the N/P ratio is less than 5, with co-
limitation likely when the ratio is between 5 and 20, and that phosphorus could cause
algal growth limitations at N/P ratios greater than 20. The N/P ratio does not directly
affect algal productivity. Rather it only identifies the nutrient that could be limiting
growth.
Carlson Trophic State Indices (TSI)
A lake's trophic condition is a measure of abiotic and biotic relationships. Carlson
(1977) and Havens (1995) introduced a set of lake trophic state indices (TSI), which use
algal biomass as the basis for trophic state classification and reflects a continuum of
"states" based on a base-2 logarithmic transformation of Secchi Disk Depth (SD) in
meters, chlorophyll-a (CHL) in ug/l, total phosphorus (TP) in ug/l and total nitrogen (TN)
in mg/l:
TSI(SD) =10*(6-(lnSD/ln2))
TSI(CHL) = 10*(6-((2.04-(0.68*lnCHL))/ln2))
TSI(TP) = 10*(6-((ln(48/TP))/ln2))
TSI(TN) = 10*(6-((ln(1.47/TN))/ln2))
Calculated TSI values represent continuum from 1 to 100, with a value of "1"
representing the "least productive" possible condition and "100" representing the "most
productive" possible condition. Because of the logarithmic scale, a doubling in the
response occurs at each 10 unit increase of the TSI value. For example, a TSI(CHL) of
32 would have three orders of magnitude less productivity than a TSI(CHL) of 62. Of
the four TSI indices listed above, chlorophyll-a (i.e. TSI(CHL)) is the index of choice for
representing the trophic state of the lake because chlorophyll best reflects the actual
amount of algal biomass in the water (Carlson 1983).
TSI Graphical Method to Identify Limiting Factors on Lake Productivity
As a general model, the TSIs adequately describe the relationship between nutrient
availability, algal productivity, and lake transparency. In theory, TSI values should result
in the same value regardless of which measurement is used, but in practice, TSI values
vary. Fortunately, these metrics often vary in predictable ways that provide insight into
chemical and biological features that are unique for each lake (Carlson and Havens
49
-------
Region 10 Lakes Assessment
2005). For example, Secchi depth will not correlate well with chlorophyll if the major
light scattering or attenuation substance in the water were clay particles or dissolved
humic color, nor would TP correlate well with chlorophyll where phosphorus was not
limiting algal growth. Deviations of the indices can therefore be used to identify such
situations (Carlson 1991).
Accordingly, Carlson and Havens (1995) proposed a Graphical Method to identify
limiting factors on lake algal biomass based on a comparison between the TSI indices.
The basis for this method is the premise that an observed deviation of chlorophyll TSI
from the nutrient TSI indicate the magnitude of nutrient limitation, while deviations of
TSI(CHL) from TSI(SD) indicates the degree of light penetration relative to the number
and size of sestonic particles (Figure A-1). Specifically, positive values associated with
the y-axis of this Figure (i.e., "TSI(CHL)-TSI(TP)") indicated a potential nutrient limiting
condition, and negative numbers indicate a potential nutrient surplus condition.
Similarly, positive values associated with the x-axis of this Figure (i.e., 'TSI(CHL)-
TSI(SD)") indicated that the algae are packaged in large colonies resulting in greater
transparencies than expected, and negative numbers indicate that light is scattered or
absorbed by very small particles, such as suspended clays or by dissolved solids.
Figure A-1. Example of possible interpretation of deviation in Trophic State Index (TSI) values using the
Graphical Method.
Increasing
Nutrient
Limitation
Increasing
Nutrient
Surplus
• Nutrient Limited
• Dissolved Color
• Clay Particles
> Nutrient Surplus
• Non-Algal Turbidity
• Nutrient Limited
• Large Algae
• High Relative Clarity
75
• Nutrient Surplus
• Zooplankton Grazing
•High Relative Clarity
TSI(CHL)-TSI(SD)
Decreasing
Relative
Clarity
Increasing
Relative
Clarity
50
-------
Region 10 Lakes Assessment
Results
Measured nutrient levels in the "inference" NLA lakes showed that primary productivity
was primarily influenced by phosphorus concentrations (Figure A-2). Specifically, algal
productivity was less than expected when phosphorus concentrations were greater than
the Michaelis-Menten saturation cutoff of 40ug/l. In addition, most lakes with water
concentrations below this limit showed a high level of potential phosphorus limitations.
Alternatively, the bottom image in Figure A-2 illustrates that there was not a discernable
pattern of algal productivity associated with nitrogen concentrations, indicating that
nitrogen is less influential factor affecting lake productivity.
Similarly, calculated nitrogen to phosphorus ratios (N/P) indicated a strong productivity
response to water column phosphorus concentrations (Figure A-3). For example, when
N/P ratio was greater than 20, most lakes had a TSI(CHL)-TSI(TP) value of greater than
zero (Indicating phosphorus limitations for these lakes). Lakes with a N/P ratio less
than 5 clearly showed that nitrogen was limiting (i.e., TSI(CHL)-TSI(TP) <0). The
bottom image associated with Figure A-3 illustrates that nitrogen did not have a clear
relationship between productivity (i.e., y axis) and the N/P ratio, indicating that nitrogen
is less influential factor affecting lake productivity.
Figure A-4 indicates that primary productivity within many of the lakes (those located in
the top left and right corners of the figure) is potentially limited by phosphorus. In
addition, this figure shows that many of the "man-made" lakes (those created through
dams or impoundments) had reduced productivity levels resulting from high turbidity
levels. Examples are the lakes located in the bottom left corner of the figure.
The bottom image in Figure A-4 illustrates these lakes grouped into Michaelis-Menten
half-saturation classes. The group called "P < 5 times Michaelis-Menten half-saturation
constant" (shown by blue diamonds in the figure) was the most centrally located around
the origin of this figure (i.e., 0,0), indicating a close relationship between observed and
expected productivity. This result could be expected because phosphorus is present at
potentially limited concentrations and the model used to evaluate these lakes (i.e., TSI
Difference Graph) was derived for phosphorus conditions. Similarly, lakes associated
with the group called "N&P < 5 times Michaelis-Menten half-saturation constant" (shown
by green squares in the figure) were located around the origin. However, they were
much more scattered around the origin, indicating less precision in the model. The one
site associated with the group called "N < 5 times Michaelis-Menten half-saturation
constant" indicated that productivity levels were much lower than expected based on
phosphorus concentrations, a result that would be expected for a nitrogen limited
system.
Alternatively, lakes associated with the group named ""Excessive" N&P, TSI(CHL)-
TSI(TP)<-10) (shown by red circles in the figure) were very deviant from the origin of the
figure, with most lakes located in the bottom left quadrant of the figure. This result
indicates that, despite high levels of nutrients, the primary production response (i.e.,
chlorophyll) is less than expected, and that the water column transparency is higher
51
-------
Region 10 Lakes Assessment
than expected based on level of observed chlorophyll in the water, indicating that non-
algal turbidity is reducing potential primary productivity in these lakes. Lakes
associated with the group named ""Excessive" N&P, TSI(CHL)-TSI(TP)>-10) (shown by
orange triangles in the figure) were shown to have productivity levels slightly below
expected conditions based on measured phosphorus concentrations.
As mentioned in the main body of the document, a phytoplankton-zooplankton observed
versus expected index (0/E index) was developed for the NLA lakes. This type of index
estimates biological condition by measuring the agreement between the expected
taxonomic composition under reference conditions with that observed at the individual
lakes, thus expressing taxa loss. Comparing 0/E results of the 90 Region 10 sample
lakes to the 11 hand-picked reference lakes, measured 0/E index conditions were
statistically (p=0.01) less than reference conditions for lakes with "excessive" levels of
both nitrogen and phosphorus (i.e., concentration exceed five times the respective
Michaelis-Menten half-saturation constants) (Figure A-5).
Recall that lake productivity for lakes associated with the group ""Excessive" N&P,
TSI(CHL)-TSI(TP)>-10" (the orange triangles in Figure A-4) was at expected levels
based on nutrient concentrations, indicating that the low 0/E index values for these
sites is a result of high productivity. Alternatively, lakes associated with the group
""Excessive" N&P, TSI(CHL)-TSI(TP)<-10" had very suppressed primary productivity. It
can be concluded that low 0/E index results for these sites is a result of high non-algal
turbidity levels. Finally, biological conditions associated with other three groups were
not statistically different than reference conditions, indicating that nutrient limitation
could be occurring within these lakes.
Summary
• Despite high observed nitrogen concentrations in R10 "inference" lakes (see
Figure 8), phosphorus is the nutrient shown to be the primary nutrient influencing
lake productivity (see Figures A-2 and A-3).
• Algal growth was lower than expected in many lakes with high nitrogen and
phosphorus concentrations, potentially indicating the effects of high turbidity (see
Figure A-4). This conclusion applies primarily to sites at man-made lakes.
• The biological response (i.e., 0/E index) was statistically lower for lakes with
"excessive" nutrient levels (see Figure A-5). This trend was not observed for
lakes with one or both nutrients below a threshold concentration (i.e., 5 times the
Michaelis-Menten half-saturation concentration).
• Low 0/E index values appear to be a results of either high productivity (for the
group named "Excessive" N&P, TSI(CHL)-TSI(TP)>-10) or high turbidity (for the
group named "Excessive" N&P, TSI(CHL)-TSI(TP)<-10).
52
-------
Region 10 Lakes Assessment
Figure A2. Potential nutrient limitations and nutrient concentrations for R10 NLA lakes.
[Positive numbers on the y-axis indicate potential nutrient limitations, and negative numbers indicate
potential nutrient surplus. The dashed red line represents potential nutrient surplus concentrations based
on five (5) times Michaelis-Menten half-saturation concentrations (i.e., 40 and 125 ug/l for phosphorus
and nitrogen, respectively.)]
Positive numbers indicate Potential Nutrient Limitations
(i.e. Greater chlorophyll-a response than expected based on nutrient concentrations)
20
10 '
• Natural Lakes
A Man Made Lakes
o
-10 -
-20 -
-30 -
-40 -
Negative Numbers indicate Potential Nutrient Surplus
(i.e. Less chlorophyll-a response than expected based on nutrient concentrations)
A
TP (ug/l)
50
40 -
30 A
20-
10 -
o
55-10
-20 -
-40 -
-50
Positive numbers indicate Potential Nutrient Limitations
(i.e. Greater chlorophyll-a response than expected based on nutrient concentrations)
• Natural Lakes
A Man Made Lakes
Negative numbers indicate Potential Nutrient Surplus
(i.e. Less chlorophyll-a response than expected based on nutrient concentrations)
TN (ug/l)
200 A 300 400 500 600 700 800 900 1000
500 1000 1500 2000 2500 3000 3500 4000 4500 50)0
53
-------
Region 10 Lakes Assessment
Figure A-3. Potential nutrient limitations and N/P ratios for Region 10 NLA lakes.
[Positive numbers on the y-axis indicate potential nutrient limitations, and negative numbers indicate
potential nutrient surplus. The left dashed red line represents the approximate N/P ratio cut-off below
which nitrogen is the limiting nutrient, and the right dashed line represents the N/P ratio above which
phosphorus is a limiting nutrient. The area between these two dashed lines represents the N/P ratio
where either nitrogen or phosphorus could be limiting algal growth.]
50
40 -
30 -
20 -
10 -
-10 -
-20 -
-30 -
-40 -
-50
50
Positive numbers indicate Potential Nutrient Limitations
(i.e. Greater chlorophyll-a response than expected based on nutrient concentrations)
*•
20 JO
1
50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 2(0
• Natural Lakes
A Man Made Lakes
Negative numbers indicate Potential Nutrient Surplus
(i.e. Less chlorophyll-a response than expected based on nutrient concentrations)
NP Ratio
40 -
30 -
20 -
10 -
55 -io-
-20 -
-40 -
-50
• Positive numbers indicate Potential Nutrient Limitations
(i.e. Greater chlorophyll-a response than expected based on nutrient concentrations)
4- A,
•^ »" ^
\ _• • u
1
0 A2D
• Natural Lakes
A Man Made Lakes
n D,
*0 40
•
60 70 80 90 100 110 120 130 140 150 160 170 180 190 2(0
Negative numbers indicate Potential Nutrient Surplus
(i.e. Less chlorophyll-a response than expected based on nutrient concentrations)
NP Ratio
54
-------
Region 10 Lakes Assessment
Figure A-4. TSI Difference Graph for Region 10 NLA lakes.
[Positive numbers on the y-axis indicate potential nutrient limitations, and negative numbers indicate
potential nutrient surplus. Positive numbers on the x-axis occurs indicates that chlorophyll is packaged in
large filamentous or colonial groups, and negative numbers indicate potential non-algal turbidity.]
> Nutrient Limited
> Dissolved Color
• Clay Particles
I -100
O,
V)
-75
-50
' Nutrient Surplus
• Non-Algal Turbidity
CL
I
X
O,
w
100 -i
75 -
50 -
I Natural Lakes
A Man Made Lakes
1 Nutrient Limited
1 Large Algae
• High Relative Clarity
-50 -
A
-75 -
25
50 75
• Nutrient Surplus
• Zooplankton Grazing
• High Relative Clarity
100
TSI(CHL)-TSI(SD)
75 -
50 -
• Nutrient Limited
• Dissolved Color D ,,
• Clay Particles
D
D<^J$
00 -75 -50 £5 L. i
• Nutrient Surplus
•.
•
• -75-
n N&P < 5 times Michael is-Menten half -saturation constants
o P < 5 times Michaelis-Menten half -saturation constants
• N < 5 times Michaelis-Menten half-saturat'on constants
• "Excessive" N&P, and TSI(CHL)-TSI(TP) < -10
A "Excessive" N&P, and TSI(CHL)-TSI(TP) > -10
0 • Nutrient Limited
• Large Algae
^<$>0 • High Relative Clarity
hx^A^
tfjf-
AAA 25 50 75 1
• Nutrient Surplus
• Zooplankton Grazing
• High Relative Clarity
TSI(CHL)-TSI(SD)
55
-------
Region 10 Lakes Assessment
Figure A-5. Comparison between calculated Zooplankton and Phytoplankton Observed/Expected Values
(O/E) and TSI Analysis Groups for Region 10 NLA lakes.
\.£.
4
(ton O/E
3
0
3hotoplanl-
3 C
r> b
kton and F
3 C
>. C
Zooplan
3 C
3 'i
\}.£-
0.
\/ \
/\ /
n=46 n= 1
\/
/
\
n=14
^ X
\/
^
N&P < 5 times P<5 times N<5 times "Excessive" N&P, "Excessive" N&P, Reference Lakes
Michaelis-Menten Michaelis-Menten Michaelis-Menten and TSI(CHL)- andTSI(CHL)-
half-saturation half-saturation half-saturation TSI(TP)<-10 TSI(TP)>-10
constants constants constants
56
-------
Region 10 Lakes Assessment
References for Appendix 3.
Carlson, R.E., 1977.A trophic state index for lakes. Limnol. Oceanogr. 22: 361-369.
Carlson, R.E., 1983. Discussion, 'Using Differences Among Carlson's Trophic State
Index Values in Regional Water Quality Assessment,' Water Res. Bull, 19: 307-308.
Carlson, R.E., 1991. Expanding the trophic state concept to identify non-nutrient limited
lakes and reservoirs. In Taggart, J. (ed.), Enhancing the States' Lake Management
Program. NE Illinois Planning Commission, Chicago, IL.
Carlson, R.E. and K.E. Havens, 2005. Simple Graphical Method for the Interpretation of
Relationships Between Trophic State Variables. Lake and Reservoir Management, 21.
Chapra, S.C., 1997. Surface Water-Quality Modeling. McGraw-Hill.
EPA/600/3-85/040, 1985. EPA Document - Rates, Constants and Kinetics
Formulations in Surface Water Quality Modeling (Second Edition).
Havens, K.E., 1995. Secondary nitrogen limitation in a subtropical lake impacted by
non-point source agricultural pollution. Environ. Poll. 89: 241-246.
Thomann R.V. & J.A. Mueller, 1987. Principles of Surface Water Quality Modeling and
Control. Harper & Row.
57
-------
Region 10 Lakes Assessment
Appendix 4. Fish Tissue—Region 10 results from 2004
National Lakes Fish Tissue Survey
Overview
The National Lake Fish Tissue Study was completed in 2004. This was the first time
contaminants in lake fish tissue were estimated across the nation. The study assessed
fish from natural and man-made lakes in the lower 48 states for the following purposes:
1) to estimate the distribution of persistent, bioaccumulative, and toxic chemical
residues in freshwater fish tissue and 2) to define a national baseline for assessing
progress of pollution control activities.
Sample lakes were selected using an uneven random design similar to that of the
National Lakes Survey (Olsen et al. 2009). Sample lakes were drawn from six size-
categories ranging from 1 to > 5,000 ha with varying probabilities of selection. The
sample design ensured the sampling of rare (i.e. large) lake size classes and the spatial
distribution of sites across states.
Fish tissue was collected from 500 lakes in the summers of 2000-2003 by the EPA and
state agencies following field protocol and quality assurance directives developed for
the survey (USEPA 2002, USEPA 2004). At each lake, crews attempted to collect two
fish species composites of 5 similar-sized adults per site: one for the human health
endpoint -predator species analyzed as fillets and one for the ecological endpoint
analysis - bottom-dweller analyzed as whole fish tissue. The study analyzed fish tissue
for 268 chemicals:
• 2 metals (mercury and 5 forms of arsenic)
• 17 dioxins and furans
• 159 PCB measurements (out of 209 congeners)
• 46 pesticides
• 40 other semi volatile organics (e.g., phenols)
Results for several important chemicals are in Stahl et al. 2009 and comprehensive
information on the survey is available at
http://water.epa.gov/scitech/swguidance/fishshellfish/techguidance/study/index.cfm.
Region 10 Fish Tissue Study
The Region 10 portion of the national lakes survey consisted of 30 sample lakes in the
three contiguous states of Region 10 (excludes Alaska). Sample lakes were distributed
across the region (Map 1) with seven in Idaho, nine in Oregon and 14 in Washington.
Sites were distributed among all lake size categories of the survey design but most
were over 500 ha in surface area (Figure A1). The sample also represents a range of
elevations with both large and small lakes distributed in range of elevations (Figure A2).
This set of 30 lakes across Region 10 resulted in a very diverse group of sample lakes
58
-------
Region 10 Lakes Assessment
ranging from pristine high elevation lakes to low elevation working reservoirs. The lakes
included in the Region 10 sample are listed at the end of this Appendix.
Map A1. Locations of 30 lake samples in Region 10.
Lakes sample by size (ha)
>5000
500-5000
50-500
10-50
5-10
1-5
2
2
2
6
7
11
0 2 4 6 8 10 12
Lakes count
Figure A1. Count of lakes sampled by size category.
59
-------
Region 10 Lakes Assessment
£
(0
IUUUUU
•tnnnn
•tnnn
mn
1
I
',
* % •• •
: •
L* " • *
T •
r •
0 2000 4000 6000 8000
Watershed Mean Elevation (ft)
Figure A2. Distribution of lakes by size and elevation.
10000
Each of the 30 lakes was fished for both a predator and bottom-dwelling species
Samples were successfully collected at 28 of the lakes for predators and at 19 of the
lakes for bottom-dwellers. We opted not to extrapolate these results to a greater
inference population because of this small sample size and the great difference in
'weights' among lakes (see Lake List table). The data presented thus expresses only
the results of these sample lakes. We report the results of total mercury, total PCBs
(the sum of the congeners analyzed), and total DDTs (sum of DDTs, DDEs, and DDDs).
These are important pollutants present in fish tissue in Region 10 as well as nation-
wide.
• Mercury is an elemental metal that is toxic at low levels, affecting the nervous
system and brain. The methylated form bioaccumulates in the food chain.
Atmospheric deposition is the largest source of mercury in the environment
(84%). Other basin scale sources are runoff, point discharges, and mining.
•
• DDT: Organochlorine pesticides including DDT were widely used in agriculture of
the Columbia Basin. DDT is highly persistent in the environment,
bioaccumulates in the food chain, and is linked to neurological and
developmental disorders in birds and other animals. It was banned in 1972, but
DDT and its breakdown products (ODD, DDE) still persist in the environment.
•
• PCBs: Polychlorinated biphenyls are synthetic compounds that were widely used
in electrical equipment. These persistent, hydrophobic chemicals
bioaccumulate in body fat and biomagnify in the food chain. PCBs have many
congeners and vary in degree of toxicity. PCB manufacture was banned in
1979 as they are carcinogenic and pose environmental and human health risk.
Data were evaluated by comparing results to screening thresholds from the literature
(Table A1). The human health threshold for mercury is a tissue based Water Quality
Criterion (USEPA 2001). Threshold values for PCBs and DDTs are from EPA's
60
-------
Region 10 Lakes Assessment
guidance for assessing chemical contaminants and are risk-based thresholds (USEPA
2000). The ecological health screening values are wildlife toxicity thresholds based on
risk to bald eagles and mink as cited in Hinck et al. 2008.
Table A1. Summary of screening value exceedances.
Chemical
Mercury
PCBs
DDTs
Human Health
Screening Value*
300 ppb
12ppb
69 ppb
Ecological Health
Screening Value**
100-300 ppb
11 0-480 ppb
150-3,000 ppb
*from EPA national reporting
** from Hinck et al. 2008
Results
Summaries of results are presented as cumulative percent graphs with the threshold
exceedance value shown in red for predators (human health endpoint) and bottom-
dweller samples (ecological endpoint). Summary statistics are in the following table
(Table A2)
Table A2. Summary statistics for fish tissue contaminants (units in ppb).
Variable
Mean
Std.
Err.
Median
Std.
Dev.
Skew.
Range
Min.
Max.
N
Predators
Mercury
Total PCBs
Total DDTs
198.18
7.01
62.84
34.85
2.20
52.62
133.00
4.18
3.48
184.39
11.64
278.44
1.30
135.68
5.27
577
57
1481
80
19
40
23.20
.533
0.00
601.00
57.72
1481.40
28
28
28
Bottom-dwellers
Mercury
Total PCBs
Total DDTs
153.92
27.02
151.65
33.01
6.54
62.61
93.60
13.33
10.49
143.90
28.51
272.92
1.86
.85
2.20
581
83
955
50
27
91
14.50
.794
0.00
596.00
84.07
955.91
19
19
19
The three types of contaminants were commonly detected in the regional samples
(Table A3). As with the National Tissue Survey results, mercury and PCBs were
detected in 100% of the samples for both predators and bottom-dwellers. The region
was consistent with the national detection for DDT as well.
61
-------
Region 10 Lakes Assessment
Table A3. Percent samples above detection limits.
Chemical
Mercury
Total PCBs
Total DDTs
Predators
100%
100%
83% (78%
nationwide)
Bottom Dwellers
100%
100%
90%
(98%nationwide)
Mercury
• Mercury levels were generally higher in predator fillet than in bottom-dweller
whole fish composites.
• The human health screening level of SOOppb was exceeded in 20% (6) of
predator composites.
• The low range of ecological endpoint screening value of 10Oppb was exceeded
in 50% (9) of bottom-dweller composites.
Mercury in Predators
100 200 300 400 500
Hg (ppb wet wt.)
600 700
100
Mercury in Bottom-dwellers
100 200 300 400 500
Hg (ppb wet wt.)
600 700
PCBs
PCBs were detected in all composites but levels were substantially higher in
bottom-dweller whole fish than in predator fillet composites.
The human health screening level of 12ppb was exceeded in 7% (2) of the
predator composites.
The ecological screening level of 110ppb was not exceeded.
Fish of both categories had more PCBs in the larger lakes.
62
-------
Region 10 Lakes Assessment
100
PCBs in Predators
20
40 60 80
PCBs (ppb wet wt.)
100
100
O
PCBs in Bottom-dwellers
20 40 60 80
PCBs (ppb wet wt.)
100
DDTs
Except for one lake, DDTs in predator composites were below the human health
screening level.
The one outlier sample with very high levels of DDTs for predators was lake
Chelan in eastern Washington. This is an area of intense long-term agriculture
where accumulation of these legacy pollutants would not be surprising.
DDTs in bottom-dweller whole-fish composites were generally higher than the
predator fillet samples.
The lower ecological screening value of 150 ppb was exceeded in 32% (6) of the
bottom-dweller composites.
Levels in neither sample type were substantially related to lake size.
100
Total DDT in Predators
20 30 40
DDT (ppb wet wt.)
50
60
100
80
> 60
40
20
Total DDT in Bottom-dwellers
200
400 600
DDT (ppb wet wt.)
800
1000
Of the three types of contaminants, the human health threshold for mercury was
exceeded for the most lakes in the Region 10 sample (Table A4). In general, the lakes
sampled in the Region were in somewhat better condition based on human health
thresholds for these contaminants compared to the nationwide results.
63
-------
Region 10 Lakes Assessment
Table A4. Percent of lakes exceeding human health thresholds.
Chemical
Mercury
Total PCBs
Total DDTs
Region 10
20%
7%
<4%
National Survey
49%
17%
2%
Recommendations:
The following are recommendations for future national-scale fish tissue studies:
• Combine the effort needed to collect fish for a national fish tissue study with the
efforts of the National Lakes Survey. The primary areas of overlap would be
sample design and lake selection/evaluation. Collection of fish can be very time
consuming, plus it requires specialized equipment and expertise. This task could
not simply be added to the sample day for Lake Survey crews.
•
• Add analyses for selenium and zinc, which would be very useful for fish tissue
monitoring in the western states.
•
• Increase sample size across the range of lake sizes. This is necessary to ensure
the ability to make inferences to the greater population of lakes at a regional
scale with reasonable statistical confidence.
64
-------
Region 10 Lakes Assessment
References for Appendix 4.
Hinck, J. E., C. J. Schmitt, et al. (2008). Environmental contaminants in freshwater fish
and their risk to piscivorous wildlife based on a national monitoring program.
Environmental Monitoring and Assessment. DOI 10.1007/s10661-008-0331-5.
Olsen, A.R., B.D. Snyder, L.L. Stahl, and J.L. Pitt. 2009. Survey design for lakes and
reservoirs in the United States to assess contaminants in fish tissue. Environmental
Monitoring and Assessment. 150:91 -100.
Stahl, L.L., B.D. Snyder, A.R. Olsen, and J.L. Pitt. 2009. Contaminants in fish tissue
from U.S. lakes and reservoirs: a national probabilistic study. Environmental Monitoring
and Assessment. 150:3-19.
US Environmental Protection Agency (USEPA). 2000. Guidance for assessing
chemical contaminant data for use in fish advisories, volume 2: Risk assessment and
fish consumption limits, 3rd ed. US EPA, Office of Water, Office of Science and
Technology, Washington, D.C. EPA 823-B-00-008.
USEPA (U.S. Environmental Protection Agency). 2001. Water quality criterion for the
protection of human health: methyl mercury. US EPA, Office of Water, Office of
Science and Technology, Washington, D.C. EPA-823-R-01-001.
USEPA (U.S. Environmental Protection Agency). 2002. Field Sampling Plan for the
National Study of Chemical Residues in Lake Fish Tissue. Office of Science and
Technology, EPA-823-R-02-004. USEPA, Washington, D.C.
USEPA (U.S. Environmental Protection Agency). 2004. Quality Assurance Report for
the National Study of Chemical Residues in Lake Fish Tissue: Year 1, Year 2, and Year
3 Analytical Data. Office of Water and Office of Science and Technology, Prepared by
DynCorp Environmental MOBIS Contract No. GS-23F-9820, Task68-C-00-137.
USEPA, Washington, D.C.
65
-------
Region 10 Lakes Assessment
Table A4. Sites sampled in EPA Region 10 as part of National Lakes Fish Tissue Survey 2000-2004.
SitelD
OWOW99-0079
OWOW99-0127
OWOW99-0554
OWOW99-0627
OWOW99-0904
OWOW99-1028
OWOW99-1452
OWOW99-0076
OWOW99-0326
OWOW99-0451
OWOW99-0629
OWOW99-0901
OWOW99-1001
OWOW99-1353
OWOW99-1454
OWOW99-1501
OWOW99-0004
OWOW99-0179
OWOW99-0202
OWOW99-0279
OWOW99-0304
OWOW99-0504
OWOW99-0529
OWOW99-0654
OWOW99-0979
OWOW99-1054
OWOW99-1354
OWOW99-1379
OWOW99-1479
OWOW99-1554
Lake Name
Brownlee Reservoir
Palisades Reservoir
Priest Lake
Bear Lake
LoonCr. Lk#1
Enoslk#1
Blackfoot Reservoir
noname gravel pit
Malheur lake
Crater Lake
Lake Umatilla
Elk Lake
Denley Reservoir
Ik Owyhee, elbow
Barney Reservoir
Wickiup Reservoir
Keechelus Lake
Frenchman Hills Ik
Cresent Lake
Nahwatzel Lake
Patterson Lake
Lake Chelan
Rimrock Lake
Lake Dorothy
Lone Lake
Potholes Reservoir
Pend Oreille River
Buffalo Lake
Lake Wallula
Calligan Lake
State
ID
ID
ID
ID
ID
ID
ID
OR
OR
OR
OR
OR
OR
OR
OR
OR
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
Longitude
(DD)
-117.0784
-111.1113
-116.8576
-111.3329
-115.9208
-115.8469
-111.5860
-123.2389
-118.7936
-122.0948
-120.5315
-122.1189
-123.2441
-117.3510
-123.3889
-121.7221
-121.3595
-119.5883
-123.7674
-123.3324
-120.2445
-120.3321
-121.1618
-121.3833
-122.4597
-119.3222
-117.2925
-118.8874
-118.9817
-121.6659
Latitude
(DD)
44.6758
43.2436
48.5679
42.0037
45.0938
45.0996
42.9042
44.5527
43.3098
42.9494
45.7258
44.8230
43.3729
43.4992
45.4452
43.6916
47.3342
46.9819
48.0848
47.2432
48.4589
48.0261
46.6403
47.5843
48.0215
46.9868
48.4300
48.0631
46.0048
47.6052
County
WASHINGTON
BONNEVILLE
BONNER
VALLEY
VALLEY
CARIBOU
LINN
HARNEY
KLAMATH
KLICKITAT
MARION
DOUGLAS
MALHEUR
WASHINGTON
DESCHUTES
KITTITAS
GRANT
CLALLAM
MASON
OKANOGAN
CHELAN
YAKIMA
KING
ISLAND
GRANT
PEND OREILLE
OKANOGAN
BENTON
KING
Design
Wgt.
1.97
1.97
1.97
1.97
904.43
904.43
1.97
233.14
1.97
1.97
1.97
236.78
233.14
9.74
72.44
9.74
9.74
72.44
9.74
72.44
72.44
1.97
9.74
72.44
236.78
1.97
9.74
72.44
1.97
72.44
Sample
Year
2000
2000
2000
2000
2002
2002
2002
2002
2003
2001
2002
2002
2002
2003
2003
2003
2001
1999
1999
2003
2003
2000
2000
2000
2001
2001
2002
2002
2002
2002
Lake Area
(ha)
6070.5
6061 .6
9453.8
28329.0
2.6
3.0
6475.2
7.2
5961.7
5318.0
11697.9
26.0
5.9
4576.9
81.1
4110.4
955.4
138.3
1995.2
111.2
51.6
13091.0
952.0
101.9
34.2
11333.0
935.8
226.2
12960.9
117.0
66
------- |