ArticlePDF Available

Abstract and Figures

Ecosystem health metrics quantify the cumulative effects of stressors on ecosystem structure and function, and inform management, restoration, and policy decisions. Freshwater ecosystems, in particular, face numerous stressors, and as a result, there is an increasing array of health metrics applied to their management. In this chapter, we review the current use of ecosystem health metrics, develop a preliminary framework for metric selection, and identify gaps in the current suite of metrics. The existing metrics typically characterize the biological, physical, or chemical attributes of ecosystems, whereas a few additional metrics integrate across these categories. Metrics vary in complexity, ranging from simple, visual assessments that can be completed by volunteers, to complex numerical models with extensive data and expert input requirements. Overall, ecosystem health metrics are well developed and useful with metrics available to fit both general and specialized management needs. However, common challenges include difficulty in establishing suitable reference conditions, a lack of uncertainty estimates, and a lack of inter-metric comparisons. Recent technological improvements, such as remote sensing, computational models, and new genetic sequencing techniques, are facilitating the development of novel and more holistic metrics, including early warning metrics, coupled complex systems models, and the inclusion of public input data.
Content may be subject to copyright.
101
Section 1. Introduction
Freshwater ecosystems face numerous stressors, includ-
ing chemical pollutants, eutrophication, loss of biodiversity,
and altered flow regimes (e.g., Carpenter et al. 1998; Milly et
al. 2005; Diaz and Rosenberg 2008; Holtgrieve et al. 2011).
Such stressors often negatively affect ecosystem structure and
function, resulting in harmful algal blooms, proliferation of
invasive species, declines in sensitive species, and reduced
water quality. Whereas stressor effects on ecosystem structure
and function are obvious in some cases, quantification of these
effects is important to compare across systems and to iden-
tify sources of impairment. The quantification of ecosystem
responses to stressors has been formalized through metrics
that describe ecosystem health. These metrics have been used
in a management context to set priorities for restoration and
conservation, delineate impaired ecosystems, identify viola-
tions of clean water legislation, and to track success of water
quality improvement efforts.
The concept of ecosystem health is popular because it is
easy to understand metaphorically, yet from a practical stand-
point, its definition remains somewhat vague. Consequently,
a clear and defensible definition of ecosystem health is crucial
for its effective application to the management of freshwater
ecosystems. Whereas “ecosystem health” has been defined in
numerous ways (e.g., Regier 1993; Karr and Chu 1999; Lackey
2001; Vugteveen et al. 2006), in this chapter we use the defi-
nition from Meyer (1997): A healthy ecosystem is “sustainable
and resilient, maintaining its ecological structure and function
over time while continuing to meet societal needs and expecta-
tions.” We chose this definition because it includes ecosystem
integrity, which is what most metrics attempt to quantify, and
because it recognizes human needs and values in managing
ecosystems.
Taking the pulse of the ecosystem: progress in quantifying aquatic
ecosystem health
Sarah S. Roley1*, Jennifer R. Griffiths2, Peter S. Levi3, Christopher J. Patrick4, Steven Sadro5, and Jay P. Zarnetske6
1W.K. Kellogg Biological Station, Michigan State University, 3700 E Gull Lake Dr, Hickory Corners, MI, USA
2Department of Ecology, Environment and Plant Sciences, Stockholm University, SE-106 91, Stockholm, Sweden
3Center for Limnology, University of Wisconsin-Madison, 680 North Park Street, Madison, WI, USA
4Smithsonian Environmental Research Center, 647 Contees Wharf Road, Edgewater, MD USA
5Earth Research Institute, University of California Santa Barbara, 6832 Ellison Hall, Santa Barbara, CA USA
6Department of Geological Sciences, Michigan State University, East Lansing, Michigan, USA
Abstract
Ecosystem health metrics quantify the cumulative effects of stressors on ecosystem structure and function,
and inform management, restoration, and policy decisions. Freshwater ecosystems, in particular, face numer-
ous stressors, and as a result, there is an increasing array of health metrics applied to their management. In this
chapter, we review the current use of ecosystem health metrics, develop a preliminary framework for metric
selection, and identify gaps in the current suite of metrics. The existing metrics typically characterize the bio-
logical, physical, or chemical attributes of ecosystems, whereas a few additional metrics integrate across these
categories. Metrics vary in complexity, ranging from simple, visual assessments that can be completed by vol-
unteers, to complex numerical models with extensive data and expert input requirements. Overall, ecosystem
health metrics are well developed and useful with metrics available to fit both general and specialized manage-
ment needs. However, common challenges include difficulty in establishing suitable reference conditions, a lack
of uncertainty estimates, and a lack of inter-metric comparisons. Recent technological improvements, such as
remote sensing, computational models, and new genetic sequencing techniques, are facilitating the development
of novel and more holistic metrics, including early warning metrics, coupled complex systems models, and the
inclusion of public input data.
*Corresponding author: E-mail: roleysar@msu.edu
Acknowledgements
Emily Norton Henry and two anonymous reviewers provided useful com-
ments on this manuscript.
Publication was supported by NSF award OCE08-12838 to P.F. Kemp
ISBN: 978-0-9845591-4-5, DOI: 10.4319/ecodas.2014.978-0-9845591-4-5.101
Eco-DAS X Chapter 7, 2014, 101–118
© 2014, by the Association for the Sciences of Limnology and Oceanography
Eco-DAS X
Symposium Proceedings
Roley et al. Aquatic ecosystem health metrics
102
A broad range of metrics have been used to assess eco-
system health. They focus on characterizing the integrity of
biological (e.g., Karr 1981), physical (e.g., Maddock 1999),
and chemical (e.g., USEPA 2012) components of the eco-
system (Table!1). Some metrics integrate across these broad
categories by combining physical, chemical, and biological
measurements into a single index (e.g., Carlson 1977), whereas
others use rates of ecosystem processes, such as gross primary
production and respiration, as indicators of ecosystem health
(e.g., Bunn and Davies 2000). Such a diversity of metrics and
approaches may imply that we have a good understanding of
the concept of ecosystem health and how to measure it. In
fact, while numerous metrics have been developed, we lack a
comprehensive, overarching framework for their application.
Choosing ecosystem health metrics poses many challenges.
One must consider how well the metric characterizes eco-
system integrity, how it relates to the stressors involved, and
whether it is appropriate for its intended usage—for example,
is it appropriate to make management decisions based on a
chosen metric? In particular, the application and interpre-
tation of any metric must consider sampling frequency and
location, natural variation in response metrics, the choice of
appropriate reference conditions, and the temporal and spatial
scale to which the results can be applied. Further complicat-
ing matters, the current set of ecosystem health metrics is
not well-distributed across the components and attributes of
aquatic ecosystems. For example, numerous metrics exist for
fish and macroinvertebrate communities, whereas relatively
few have been developed for ecosystem processes. As a result,
it may be unclear which metric is most appropriate (e.g., how
to choose the best fish metric), and in some cases, an appropri-
ate metric may not be available (e.g., assessment of ecosystem
function). Such challenges stem in large part from the lack of
a universal framework to select the appropriate suite of eco-
system health metrics.
Our goal with this review is to move toward a more com-
prehensive understanding of ecosystem health metrics and the
context in which they are applied. We begin by summarizing
the state of knowledge on ecosystem metrics, and then develop
a preliminary framework for metric selection by detailing
appropriate metric usage and limitations across a range of
freshwater systems. Finally, we identify gaps in the current
suite of metrics and highlight aspects of freshwater ecosystems
that are not currently characterized by any existing metrics.
This discussion centers on metrics used in North America, but
our general conclusions should be applicable across the globe.
This synthesis provides a starting point for those interested in
identifying appropriate metrics for their own systems, and also
identifies avenues for new research where metrics are lacking.
Section 2. Overview and application of biological
metrics
There is a broad suite of biological metrics used to assess
ecosystem health in freshwater systems. They include numer-
ous taxa, from bacteria to fish, and some metrics are applied
simultaneously to multiple taxonomic groups (e.g., Pont et
Table 1. Categories of ecosystem health metrics.
Category Type Examples
Relevant
ecosystems*Ease of use Limitations
Biological MMIs Index of biotic integrity
(IBI)
S, L, W Easy collection, processing can be lengthy Defining reference conditions,
linking stressors
Single-parameter Invertebrate functional
groups
S, L, W Easy and inexpensive Defining reference conditions,
linking stressors
Organism health Fish parasite or
contaminant load
S, L Specialized laboratory required Linking results to ecosystem health
Human health E. coli population S, L Specialized laboratory required Interpretation
Physical Single-parameter Secchi depth S, L, W Easy and inexpensive Defining reference conditions,
linking to stressors
Visual surveys Reach habitat assessment SEasy and inexpensive Defining reference conditions
Dynamical models MesoHABSIM S Requires specialized expertise Interpretation challenging
Chemical Nutrients N and P concentrations S, L Easy collection and processing Defining impairment
Emerging
contaminants
Caffeine, pharmaceuticals S, L Specialized laboratory required Effects on ecosystem health largely
unknown
pH, alkalinity pH S, L, W Easy and inexpensive Detects only specific stressors
Integrative Multiple metrics Trophic State Index (TSI) LEasy collection and processing Defining reference conditions
Ecosystem process Metabolism S, L Easy collection, interpretation difficult Defining reference conditions,
defining impairment
*S = streams, L = lakes, W = wetlands
Roley et al. Aquatic ecosystem health metrics
103
al. 2009), whereas others are taxon-specific (e.g., USEPA
2010, 2013). In addition, the structure of the biological met-
rics varies, and includes both multi-parameter (multivariate,
multi-metric) and single parameter indexes. Multi-parameter
metrics create an index score from multiple measurements,
whereras single parameter metrics focus on a single indi-
cator. Multivariate and multimetric indices primarily differ
in whether site classifications are determined a posteriori
(multivariate) with multivariate statistical methods (e.g., ordi-
nation, regression) or a priori (multi-metric) based on a suite
of physicochemical and community metrics (Bowman and
Somers 2005).
Many metrics are intended for a discrete area, such as
a small stream reach or wetland, but others take a broader
view and compare across sites (e.g., Soto-Galera et al. 1999).
Whereas many biological metrics have been developed, the
extent to which they have been applied across ecosystems
remains uneven. For example, more metrics exist for streams
than for lakes or wetlands. Additionally, there are other
metrics that could be expanded to include more taxa (e.g.,
contaminant loading is currently focused on fish, but may be
valid for other taxa).
Multi-metric indices (MMI) are the most extensively used
biological metric to assess ecosystem health. These indexes,
such as the Index of Biological Integrity (IBI, Karr 1981),
incorporate numerous community-level parameters to gen-
erate a holistic picture of the communities and the quality of
their environment. Examples of contributing metrics include
species richness, presence of tolerant species, proportion of
feeding guilds, and presence of rare taxa. Aquatic MMIs are
most often used in streams and rivers (Ruaro and Gubiania
2013), but have also been adapted to lakes (Beck and Hatch
2009) and wetlands (Burton et al. 1999). MMIs typically
include metrics that measure attributes relevant to a particular
ecosystem type or to account for an intrinsic ecosystem char-
acteristic (e.g., a species-poor system, Aparicio et al. 2011).
In addition, the types of measurements included in MMIs
depend on the taxonomic focus. For example, while initially
developed for fish communities, IBIs have also been developed
for macrophytes (Beck et al. 2010), plankton (Kane et al. 2009),
invertebrates (Kerans and Karr 1994), and combinations of
these groups (Pont et al. 2009). Although useful, MMIs are
often developed for relatively localized areas based upon least
disturbed or best available conditions (Bowman and Somers
2005), which complicates applying them in areas for which
they were not designed. To overcome these limitations, recent
efforts are focused on developing MMIs at regional, national,
and continental scales (e.g., Stoddard et al. 2008).
Single-parameter biological metrics include community
metrics, population parameters, multi-site comparisons, and
harmful species. Community metrics characterize functional
and taxonomic composition and are broadly used across taxa,
including algae (benthic and phytoplankton), submerged
aquatic vegetation (SAV), invertebrates, amphibians, and
fish (Cummins and Klug 1979; Attrill and Depledge 1997;
Poff et al. 2006; USEPA 2011a, Guzy et al. 2012). In addi-
tion, community metrics can focus on the proportion of the
community that is native versus non-native (Davies et al.
2010). Population parameters include biomass, abundance,
and density of an indicator species or taxon, and these met-
rics are used for algae, macrophytes, invertebrates, and fish
(Hauer and Lamberti 2007; Davies et al. 2010; USEPA 2011a).
An alternative approach is to infer ecosystem health from
fish contaminant or parasite loads (Corsi et al. 2003; Palm
and Ruckert 2009; Pietrock and Hursky 2011) or behavioral
patterns (Gorman et al. 2012). Finally, some single-parame-
ter metrics convey risk to humans via population counts of
harmful species, including algae and bacteria (e.g., Wade et al.
2006). These metrics are not typically used explicitly as indica-
tors of ecosystem health, although the proliferation of harmful
species may indicate a larger ecosystem health problem, such
as increased nutrient inputs.
Whereas most metrics are indicators of ecosystem health
at a discrete location, some metrics encompass a larger spatial
extent. These metrics address both the spatial pattern of eco-
system health and the magnitude of response to environmen-
tal conditions. Such analyses are conducive to large-bodied
organisms and were developed for SAV and fish. SAV metrics
include percent cover or habitat occupied, with the underlying
asumption that macrophyte presence reflects good ecosystem
health (USEPA 2011a). Fish metrics of spatial heterogeneity
compare current versus historic distributions within and
among watersheds, using changes in the proportion of native
fishes and pollution-tolerant fishes to make inferences about
spatial and temporal changes in ecosystem health among
water bodies within a single watershed or among watersheds
in a region (Soto-Galera et al. 1999). Contaminant load, which
can be rapidly assessed by taking samples from fish at multiple
locations, is also be used to characterize the spatial patterns of
human impacts on aquatic ecosystems (e.g., Corsi et al. 2003).
Metrics that incorporate multiple sites and spatial heteroge-
neity can allow for comparisons across a broader spatial scale,
but they are not a common metric so far, perhaps because of
their extensive data requirements or because there are fewer
metrics that explicitly incorporate these comparisons.
Biological metrics are applied across stream, lake, and wet-
land habitats, but more metrics exist for stream habitats, both
in terms of taxa use and type of metric. In general, invertebrate
and fish metrics are considered integrators of whole-ecosys-
tem condition due to their higher trophic status. In contrast,
algae, macrophytes, and zooplankton often appear to be used
for more narrowly-focused water quality and trophic state
assessments. Amphibian-based biological metrics exist, but are
less commonly implemented than metrics for other taxa (but
see Welsh and Hodgson 2008 for an example; and Kroll et al.
2009 for a critique). Biological metrics, though diverse in form,
strive to encompass multiple biological aspects of the ecosys-
tem and provide numerous ways to assess ecosystem health.
Roley et al. Aquatic ecosystem health metrics
104
Section 3. Advantages and disadvantages of
biological metrics
Biological metrics are useful indicators for a number of rea-
sons. They can be inexpensive to measure, completed rapidly
with experts or volunteers, and can provide an informative
assessment of the ecosystem. Furthermore, biological metrics
can integrate through time and can detect environmental
impacts that may not be readily detectable via physical or
chemical assessments. Finally, numerous biological metrics
exist, making it easier to find a metric suitable to specific mon-
itoring or research goals across multiple spatial scales.
Biological metrics are also advantageous because they can
be effective at monitoring the response of an ecosystem to
short-term pulse disturbances. Pulse disturbances can be diffi-
cult to detect with direct measurements, because they require
either 1) making real-time observations when the short-term
event occurs, or 2) deploying expensive long-term monitoring
equipment. However, community composition can shift in
response to short-term disturbances, with declines in sensitive
taxa and increases in tolerant taxa. In addition, many sensitive
invertebrates, fish, and amphibians live for many years, and so
their presence indicates consistently good water quality over
their long lifespans, assuming they remain within a particular
water body or a constrained stream reach.
A major strength of biological response variables are the
number of biological metrics that can be measured rapidly and
inexpensively. Rapid bioassessment protocols used by the EPA
and state agencies allow for very quick and affordable assess-
ments of stream ecosystems (USEPA 1990; Davis et al. 1996).
With a low-cost sampling device such as a kick net and a basic
understanding of aquatic invertebrate taxonomy, a team can
rapidly grade an aquatic system. Fish IBI measurements ini-
tially require more expensive equipment (e.g., nets, backpack
shockers, or shock boats), but can provide similarly efficient
assessments. In addition to collecting field data, remote sens-
ing data can provide macrophyte estimates to assess ecosystem
health, which is also inexpensive. Remote sensing data require
some ground-truthing and the appropriate skills and software
to convert the data layers into usable coverage and density data.
Because biological metrics became more popular over the
past 30 years, a number of advances have occurred, leading to
more sophisticated metrics (Karr 1981; Karr et al. 1986; Ruaro
and Guibiani 2013). For invertebrate-based metrics, many of
these advances rely on identification of invertebrates to the
lowest possible taxonomic resolution, which is often the spe-
cies level. As a result, whereas improved taxonomic resolution
can increase index reliability, it is often time-consuming and
expensive and can require specialized personnel. Similarly, the
accuracy of fish metrics improve with spatial extent surveyed,
but increasing survey area may place practical limitations
on equipment and labor. The advantages of biological met-
rics—speed and affordability—can disappear with improved
accuracy. Thus, the need for improved metric accuracy must
be weighed against budgetary constraints. The emerging field
of metagenomics, identifying suites of species based on genetic
techniques, may eventually resolve the trade-off between accu-
racy and cost.
An emerging technology that may improve biological met-
rics is the ability to detect environmental DNA (eDNA), which
is DNA that was secreted or sloughed off by organisms living
in a water body. Researchers can collect water samples, extract
DNA, and compare it with molecular markers to determine if
a target species is present (Jerde et al. 2011). This technique is
more sensitive than survey methods, because it does not rely
on physically encountering a species; DNA mixes well in the
water column, whereas organisms tend to be found only in
particular habitats (Pilliod et al. 2013). Furthermore, eDNA
eliminates some of the problems with delineating sampling
locations—while organisms may swim out of reach, eDNA
remains in the water column for up to 2 weeks (Thomsen
et al. 2012). So far, eDNA is successful in detecting certain
rare (Goldberg et al. 2011; Pilliod et al. 2013), endangered
(Thomsen et al. 2012), and non-native species (Jerde et al.
2011), but it has not yet been developed for IBIs or other
multi-species metrics. Detection of eDNA is still a new tech-
nique, and many methodological details must be worked out
before it is applied to multi-species metrics. For example, its
use in IBIs will be limited by the fact that eDNA does not, as
of yet, detect organismal abundance, just presence or absence.
Nonetheless, eDNA techniques may improve the sensitivity of
traditional survey methods, while decreasing labor and cost
(Thomsen et al. 2012).
The application of a metric to the ecosystem of interest
must also be carefully considered. In many cases, biological
metrics are developed for particular regions, and there is a risk
that a metric will perform poorly when it is used elsewhere.
Furthermore, ecosystems that differ from the idealized healthy
ecosystems on which models are based will perform poorly.
For example, stream IBIs that include fishes often consider
cold-water species, such as trout, indicative of a healthy eco-
system (Wang et al. 2003). Therefore, warm-water streams
that lack native cold-water species will perform poorly with
such a model. As with most metrics, care must be taken in the
choice, application, and interpretation of biological metrics.
Another disadvantage of biological metrics is the assump-
tion that biological integrity is closely linked with environmen-
tal quality, or that the biological community accurately reflects
the current underlying physical and chemical conditions in a
site (Karr 1981). For this assumption to be correct, the resident
community must respond instantaneously to environmental
changes, but often there is a time lag before perturbed com-
munities stabilize. Consequently, it is critical to determine
whether such time lags are large enough to affect the ability
of biotic metrics to characterize ecosystem health. The longer
the lag time in biotic response, the greater the likelihood that
sampling will occur during a transitional point in the resident
community; therefore, the greater the likelihood that the sam-
pling does not accurately reflect current ecosystem health.
Roley et al. Aquatic ecosystem health metrics
105
Meta-community ecology recently put an increased empha-
sis on the role of dispersal in community dynamics (Leibold
et al. 2004). Dispersal can interfere with the interpretation of
biological metrics in two ways. First, if dispersal limitation
occurs and is unrelated to ecosystem health, we may underes-
timate the ecosystem health due to a time lag between physi-
cochemical recovery of the system and the return of sensitive
indicator species (Patrick and Swan 2011; Bogan and Boersma
2012). The alternative mechanism, mass effects, occurs when
an unhealthy system receives inputs of sensitive species from
a nearby healthy system, artificially improving its biological
metric scores (Brown and Swan 2012). The mere presence of
highly sensitive taxa does not necessarily mean that they are
able to persist in that habitat and complete their life cycle. As
a result, scores on biological metrics should be interpreted in
light of physicochemical and mass effect conditions.
Biological metrics exist for many taxa and ecosystems, and
there are numerous options for characterizing biological eco-
system health. However, the nature of biological communities
can make these metrics difficult to interpret. In addition to the
challenge of defining reference conditions, biological metrics
must be interpreted in light of other confounding factors such
as those that affect dispersal, time-lags in response to improve-
ments in local conditions, and consideration of whether the
metric is appropriate for the region and water body type.
Overall, biological metrics provide a strong initial assessment
of ecosystem health, but must be interpreted carefully and,
ideally, within the context of an ecosystem’s physical and
chemical template.
Section 4. Physical metrics
Physical metrics include simple single-parameter mea-
surements (e.g., water clarity, Wetzel 2001), but are often
multi-parameter metrics that describe attributes of the
water and geomorphic setting of aquatic ecosystems. The
multi-parameter metrics typically combine several elements
of the physical environment, such as slope, flow, water depth,
and shoreline characteristics, to provide a single index.
Techniques vary from rapid visual inspection of flow and
geomorphic conditions (e.g., Pfankuch 1975; Rankin 1989)
to modeling of numerous system parameters (e.g., Bovee et
al. 1998; Parasiewicz and Dunbar 2001; and Milhous and
Waddle 2012).
The spatial and temporal dynamics of the physical habitat
are determined by the interaction of the structural features
and hydrological regime of the aquatic ecosystem (Maddock
1999). Therefore, the physical conditions used to assess eco-
system health often focus on quantifying the characteristics
of the flux and storage of water and sediment, as well as the
geomorphic aspects of lotic and lentic environments. In many
cases, the acceptable range of physical conditions is defined by
the documented habitat requirements for a targeted individual
species (e.g., seasonal fish habitat requirements, Gillette et
al. 2006), and thus are often used in concert with biological
metrics. In some cases, physical metrics are able to explain the
absence of certain species even if other chemical and biological
metrics indicate high ecosystem health.
Physical metrics are most commonly applied in streams,
where physical processes can have a strong effect on habitat
and ecosystem function, and where the water levels can change
rapidly enough to have a strong effect on biological responses.
They are also applied in lakes and wetlands, although fewer
lentic metrics exist, and they are generally less complex.
In streams, some physical metrics are measured by field-
based surveys that primarily draw upon visual assessments
of ecosystem conditions, including the Qualitative Habitat
Evaluation Index (QHEI), the Environmental Monitoring
and Assessment Program (EMAP West), the Pfankuch chan-
nel stability, and the Vermont Agency of Natural Resources
(VANR) Reach Habitat Assessment (RHA) (Pfankuch 1975;
Rankin 1989; VANR 2008; Stoddard et al. 2005). The VANR
RHA is a good example of a field-based survey, and it includes
rapid assessments of woody debris cover, stream bed substrate
cover, scour (pool) and depositional (riffle) geomorphic fea-
tures, reach channel morphology, flow regime characteristics,
surface flow connectivity, river banks characteristics, and
riparian area conditions (VANR 2008). In general, field-based
visual surveys provide a relatively fast way to evaluate and
combine numerous ecosystem elements into a single index.
An alternative approach is dynamical and quantitative eco-
system health assessment methods, which are emerging from
research on environmental flows (e.g., Poff et al. 1997; Dyson
et al. 2003; Arthington et al. 2006) and ecogeomorphology
(e.g., Wheaton et al. 2011). These methods couple empirical
and theoretical approaches through the development of sim-
ulation models (e.g., MesoHABSIM, Parasiewicz 2008; and
PHABSIM, Milhous and Waddle 2012). These models are
typically parameterized and calibrated with detailed field and
hydrological data, and then used to evaluate many aspects
of aquatic ecosystem health, including habitat availability.
In addition, many of these models explore how criteria for
organisms in the stream vary with changes in the flow regime,
geomorphic conditions, and habitat quality. The models are
also capable of parameter sensitivity analysis, which can iden-
tify the key controls on habitat conditions across different
hydrogeomorphic settings. In addition, dynamical models use
potential parameter distributions, rather than single fixed val-
ues or categorical scores, and determine which combinations
of parameters exceed ecosystem health thresholds (e.g., insuf-
ficient flows for a species of concern). These features of the
model are then able to help refine efforts of field sampling and
surveying. Overall, these models incorporate numerous phys-
ical attributes and predict responses to environmental change
and management, while incorporating system variability.
In lakes, physical metrics include both single- and multi-pa-
rameter indexes. The most common lake physical measure-
ment is likely the Secchi depth, an easy measurement of
water clarity that can be completed by volunteers. In general,
Roley et al. Aquatic ecosystem health metrics
106
water quality increases with Secchi depth, because stressors
such as eutrophication and erosion result in lower Secchi
depths (Bruhn and Soranno 2005). More sophisticated optical
analyses of lake waters, such as the suite of climate forcing
optical indices (CFOI) proposed by Williamson et al. (2014),
show promise as indicators of large scale climate change
effects across a wide range of lake types. In contrast, most
multi-parameter physical metrics involve visual assessments
of shoreline habitat, in which lakes with natural vegetation,
macrophytes, and natural banks score well, whereas lakes with
manicured lawns, trash, and unnatural beaches score poorly
(USEPA 2011a; McGoff et al. 2013). Often, the multi-parame-
ter metrics are used in conjunction with biological data, where
the physical metrics provide information on the physical
stressors and habitat availability. Similarly, wetland physical
metrics are often descriptive, and include soil profile and
physical characteristics, water source and depth, and hydro-
logical stressors (e.g., USEPA 2011b). They are typically used
to determine the source of stress to biota or to provide infor-
mation on the potential for wetland nutrient uptake or other
ecosystem services, rather than functioning as independent
indicators.
Section 5. Advantages and disadvantages of
physical metrics
Survey-based physical assessments all share fundamental
strengths and limitations. They can provide high-quality
and easy-to-interpret data, which can guide management
efforts, and ultimately reduce ecosystem stressors, especially
those that affect organisms. Often, survey data are easy to
obtain, and data collection can be completed by volunteers.
Furthermore, they can indicate which activities and pertur-
bations lead to long-term beneficial or harmful conditions
in aquatic ecosystems (e.g., long-term RHA assessments that
chronicle the effects of land-use change on sediment trans-
port). The physical metrics in this type of assessment also pro-
vide important information about the potential for recovery of
a degraded ecosystem upon alleviation of other biological and
chemical stressors. Therefore, they provide information about
the fundamental hydrogeomorphic template of an aquatic
ecosystem and the ability of that template to support a healthy
ecosystem.
Survey-based physical assessments also share common
limitations in their ability to evaluate ecosystem health. In
these surveys, many measurements are made of different
categories of ecosystem health, where categories are deter-
mined a priori, and these measurements are converted to
scores of overall health, based on a comparison with refer-
ence conditions. These scores may vary between practitioners
and lead to subjective sources of uncertainty. In addition,
inappropriate reference conditions can make conclusions
questionable. Last, all of these survey methods require signif-
icant extant data on targeted species or ecosystem qualities to
identify their fundamental ecosystem needs and tolerances.
The distillation of many components to a single score is also
a source of uncertainty. This uncertainty will continue to
diminish as more studies and better understanding of a larger
suite of species become available for aquatic ecosystems. Still,
survey-based physical assessments can rapidly provide useful
information on ecosystem health, as long as the results are
interpreted with care.
Flow regime is a good indicator of habitat type and biodi-
versity potential in streams (Harper et al. 2000), so measuring,
monitoring, and modeling flow and sediment conditions
are key components of evaluating stream ecosystem health
(Strange et al. 1999). The hydrogeomorphic modeling efforts
require detailed field and hydrological data to parameterize
and calibrate the models, but once the model is developed,
many aspects of aquatic ecosystem health, including habitat
availability, can be evaluated within the model system. This is
particularly advantageous over the survey-based assessment
methods, because they are capable of simulating different
conditions in the ecosystem (e.g., flow and sediment regime
changes) and determining the impact of various management
scenarios on ecosystem features, such as fish habitat (e.g.,
Parasiewicz 2008).
Aquatic ecosystem habitat models were previously limited
by computational demands and the ability to characterize and
parameterize the model domain (Maddock 1999). However,
new advances in computing and remote sensing techniques
are removing these limitations. For example, the increasingly
regular use of LIDAR (Laser Interferometry Detection and
Ranging) and UAV (unmanned aerial vehicle) remote sensing
technology in watershed surveys and river habitat mapping
will increase time-efficiency and allow for the linking of
geographic measurements across scales. Many small streams
can have their morphology and flow conditions well-charac-
terized with LIDAR at high spatial and temporal resolutions
(e.g., Wheaton et al. 2013), which makes habitat model (e.g.,
MesoHABSIM) parameterization much less difficult and
more precise than field ground-survey methods. LIDAR is a
particularly exciting technology because it is a single tool that
can map micro-habitat to watershed scales, which offers the
opportunity to integrate physical ecosystem health metrics and
processes across numerous scales. Whereas physical habitat
models require substantial investment in time and resources,
their predictive power and flexibility make them a powerful
tool. Thus, they may be particularly useful in high-profile sites
or those that have special management concerns.
Section 6. Chemical metrics of ecosystem health
Chemical metrics are based on the concentration of com-
pounds in aquatic systems, including both naturally produced
and synthesized compounds. These metrics can be divided
into three categories: nutrients and organic compounds;
human-synthesized compounds; and pH and alkalinity. The
most common chemical metrics used to assess freshwater
ecosystem health are dissolved inorganic nitrogen (DIN),
Roley et al. Aquatic ecosystem health metrics
107
including ammonium (NH4
+) and nitrate (NO3
), total nitrogen
(TN), phosphate (PO4
3–), and total phosphorus (TP). Nitrogen
and phosphorus are related to ecosystem health because they
often limit biotic production in rivers, lakes, wetlands, and
estuaries (Schindler 1977; Howarth and Marino 2006; Elser
et al. 2007). Recently, dissolved organic matter (DOM) rose
in prominence as an indicator of aquatic ecosystems health
(e.g., Roulet and Moore 2006). Together, excess nutrients and
organic matter enrichment are two of the leading causes of
freshwater impairment in the United States (USEPA 2012).
Although DIN and PO4
3– are present due to natural processes
(e.g., biogeochemical transformations, weathering), human
activities have greatly accelerated their loading into freshwa-
ters (Vitousek et al. 1997; Smith et al. 1999). Elevated nutrient
concentrations from human activities may not only affect the
freshwater ecosystems in human-dominated landscapes, but
in remote ecosystems as well (Holtgrieve et al. 2011).
Regulatory agencies establish region-specific standards
for inorganic nitrogen and phosphorus that vary in order to
account for natural variation in climate and geology (e.g.,
USEPA 1998; USEPA 2000a). For example, the United States
Environmental Protection Agency (USEPA) established cri-
teria across 14 ecoregions for lakes and reservoirs of the US
(8–38 μg L–1 of total P, 0.1–1.3 mg L–1 of total N) as well as
for streams and rivers (10–76 μg L–1 of total P, 0.2–2.2 mg L–1
of total N) (USEPA 2000a). These criteria are often highest
for the ecoregions that are most heavily impacted by humans
(e.g., Corn Belt/Northern Great Plains of the US) and lowest
in ecoregions with more forests and less agricultural land (e.g.,
Western Forested Mountains of the US) (e.g., USEPA 2000b).
Manufactured compounds in freshwaters, such as phar-
maceuticals, pesticides, and nanoparticles, can cause adverse
effects on ecosystem and human health (e.g., Schwarzenbach
et al. 2006; Griffitt et al. 2008). Along with N and P, the
USEPA lists pathogens, mercury, and heavy metals as the
primary causes of freshwater impairment in the US (USEPA
2012). Freshwater ecosystems in human-dominated land-
scapes are often the most impacted by man-made compounds
that contaminate ecosystems via sewage, industrial waste,
and intensive agricultural practices (e.g., Buerge et al. 2003).
However, man-made compounds have also been recorded in
relatively pristine ecosystems (e.g., Mast et al. 2007), which
suggests that monitoring freshwater ecosystems for such
compounds should not be restricted to human-dominated
landscapes. In addition, these compounds can be found in the
water column and stored in sediments, so their persistence
in the environment may be extensive. The ecological effects
of man-made compounds are as diverse and varied as the
compounds themselves, ranging from the relatively short-
term, innocuous effects of elevated caffeine on stream biofilms
(Lawrence et al. 2012) to the long-term, toxic effects of heavy
metals on freshwater biota (Schubert et al. 2008). Risk assess-
ments and ecotoxicology studies are becoming more com-
mon, but standards for many of these compounds are not yet
established. Rather, the presence of one or several compounds
can indicate that the health of the ecosystem is compromised
(e.g., Nakada et al. 2008).
The pH and alkalinity of freshwaters is a third category of
chemical metrics used to assess freshwater ecosystem health.
Human impacts, such as mining and industrial emissions,
can elevate the concentration of hydrogen ions (H+) in fresh-
waters via both direct and indirect inputs (e.g., acid mine
drainage and NOX and SOX emissions leading to acid rain,
respectively) (Driscoll et al. 2001). Research has demonstrated
that the biological community and overall ecosystem function
are impaired in freshwaters with unnaturally low pH (e.g.,
<5), compromising the development and/or physiology of
freshwater organisms and altering the ecosystem’s buffering
capacity (Schindler et al. 1985; McCormick et al. 1994; Likens
et al. 1996; Niyogi et al. 2002; Kaushal et al. 2013). The use of
pH as a metric of freshwater ecosystem health is often targeted
in regions where the impact of humans will most likely cause
acidification (USEPA 2005). Because of the strong relationship
between freshwater acidification and human activities (e.g.,
mining, industry, atmospheric deposition of NOX and SOX),
regulatory agencies of many countries have established guide-
lines for these activities in an effort to reduce the acidification
of freshwaters (e.g., Title IV of the 1990 amendments to the
US Clean Air Act).
Section 7. Advantages and disadvantages of existing
chemical metrics
Nutrient concentrations in freshwaters are robust metrics
for assessing freshwater ecosystem health. Whereas rapid
bioassessments may indicate ecosystem impairment, mea-
surements of inorganic N and P may indicate the potential
mechanism of the observed impairment (USEPA 1991).
Interpretation of N and P data are straightforward and the
results can be compared easily across space (within a single
water body and among water bodies regionally) and time.
However, N and P concentrations fluctuate on various time
scales due to storm events, biological uptake, watershed
characteristics, or seasonality (Welter et al. 2005; Petrone et
al. 2006). Therefore, sampling at regular intervals or under
similar conditions (e.g., baseflow in streams, seasonal turnover
in lakes) is most effective for relating these concentrations to
ecosystem health. Laboratory facilities are necessary for accu-
rate measurements, but water sampling kits are available for
coarse measurements, allowing local municipalities or citizen
scientists to assess N and P concentrations of their local fresh-
water ecosystems (Thornton and Leahy 2012).
The USEPA has established nutrient criteria for freshwaters
within the conterminous US (USEPA 2000a), which is a help-
ful starting point. However, the approach used by the USEPA
to assign nutrient criteria to each ecoregion is one of several,
each of which may suggest different criteria within an ecore-
gion (Herlihy et al. 2013). Other techniques used to establish
criteria and/or reference conditions for an ecoregion include
Roley et al. Aquatic ecosystem health metrics
108
paleolimnological reconstruction (e.g., Herlihy et al. 2013) or,
in human-dominated landscapes where few reference ecosys-
tems exist, a covariance approach among impaired ecosys-
tems (Dodds and Oakes 2004) or an ecosystem classification
approach (Soranno et al. 2010). Furthermore, the established
nutrient concentrations provide a useful guideline, but eco-
systems below these concentrations may still be impaired. For
instance, TN concentrations above 1 mg L–1 suggest human
impact on a stream or river, but these ecosystems would
not be considered impaired given the criteria established for
some ecoregions (e.g., Corn belt of the US; USEPA 2000b).
Therefore, nutrient concentrations should only be used as one
of several tools used to assess the health of an ecosystem.
The limitations of N and P as metrics of ecosystem health
are primarily a function of variability in freshwater ecosys-
tems. Whereas measuring concentrations of certain solutes is
relatively straightforward and inexpensive, simply comparing
the data to an established standard may not accurately assess
ecosystem health. For instance, previous studies demon-
strated that lakes are often P-limited and marine estuaries
N-limited (e.g., Howarth and Marino 2006), but these gener-
alizations are broad and do not apply to all lake or estuarine
ecosystems (Elser et al. 2007). Furthermore, a given freshwa-
ter ecosystem may be more or less sensitive to small changes
in the concentration of a specific solute due to the physical
and/or biological characteristics of that particular system.
Therefore, similar changes in a solute concentration may
have a disproportionately large effect in some ecosystems,
but not in others. Because broad nutrient criteria may not be
applicable to all freshwater ecosystems, more nuanced mea-
surements may provide the detailed information necessary
for the appropriate management of specific ecosystems, such
as the more integrative nutrient limitation and demand stud-
ies discussed below.
The standards established for man-made compounds are
nearly as diverse as the compounds themselves, with the stan-
dards varying according to the toxicity of that compound on
biota and its residence time in the environment (e.g., Schwab
et al. 2005). The majority of standards that do exist are estab-
lished for human, not ecosystem, health and expressed as
tolerable daily intake (TDI; USEPA 1989). However, there are
many groups of compounds present in freshwaters that likely
pose a risk to human and ecosystem health (e.g., pesticides,
such as diazinon and dieldrin), but have no TDI or similar
standards established (Murray et al. 2010). Often, the presence
of emergent pollutants alone would indicate that the health of
the ecosystem is impaired (e.g., Buerge et al. 2003), but more
research is needed to establish appropriate standards and reg-
ulation of these compounds. In general, the use of man-made
compounds as assessment tools for freshwater health may not
be practical due to the lack of standards and measurement
techniques, the variety of compounds present, the extent to
which these compounds may become buried in or remobilized
from sediments, and the highly variable concentrations of
these compounds in time and space (e.g., Kolpin et al. 2002).
Furthermore, the analysis of water or sediment samples for
many of these compounds is often costly (Schwarzenbach et
al. 2006), which makes broad temporal or spatial sampling and
monitoring impractical.
A further disadvantage for using man-made compounds as
indicators of ecosystem health is that many unknowns exist in
regards to the rapidly increasing number of these compounds
and their interactive effects. For instance, the combination
of certain pharmaceuticals in aquatic ecosystems can have
synergistic effects on ecosystem health (e.g., Cleuvers 2003).
Pharmaceuticals are often challenging to understand because
of high variation in their chemical characteristics, which may
vary for the same compound (i.e., polymorphism) or change
following human metabolism (Cunningham 2008). Though
many man-made compounds may adversely affect freshwa-
ter ecosystems, research on their effects on freshwaters may
never catch up to novel compound development (Deblonde
et al. 2011). A possible solution is to prioritize efforts toward
certain classes or groups of these compounds, where the
potential ecosystem effects may be greatest due to high toxicity
or widespread prevalence (e.g., Sanderson et al. 2004; Murray
et al. 2010). Ultimately, the development of environmentally
benign compounds will be the most effective way to minimize
the impacts of man-made compounds on ecosystem health
(Schwarzenbach et al. 2006).
Measurements of pH are a direct and straightforward
assessment of ecosystem health. The pH of freshwaters is
easy and inexpensive to measure. Low pH is often the result
of chronic impairment, although some ecosystems are natu-
rally more acidic than others. Freshwaters, in some cases, are
able to recover from unnaturally low pH (e.g., Stoddard et al.
1999), but the recovery is often over long time scales, such as
multiple years or decades. Therefore, regular monitoring may
be useful to compare across ecosystems, whereas long-term
monitoring is necessary to assess whether a system is becom-
ing impaired or in recovery.
Section 8. Integrative metrics
Integrative metrics reflect the activities of multiple eco-
system components operating in tandem, providing a more
holistic representation of ecosystem health. Because differ-
ent environmental stressors may affect ecosystem structure
or function to varying degrees, an approach that focuses on
specific aspects of structure alone may misrepresent system
health. In contrast, integrative metrics combine elements of
ecosystem structure and function. There are two approaches
to characterize ecosystem health in an integrative way. The
multi-metric approach combines various physical, chemical,
and biological measurements into a single synthetic index.
In contrast, a process-based approach uses measurements of
ecosystem processes that themselves integrate across multiple
ecosystem components. Both approaches, however, attempt to
measure functional aspects of ecosystem health directly, either
Roley et al. Aquatic ecosystem health metrics
109
by combining structural metrics associated with underlying
drivers of function, or by measuring the ecosystem function
itself more directly.
Integrative multi-metrics have been in use since the 1970s.
Perhaps the most widely used is the trophic state index
(Carlson 1977), which combines water clarity, nutrient con-
centration, and chlorophyll-a concentration to create a single
metric of ecosystem productivity and health. Multiple metrics
are also combined to reflect specific bioregional monitoring
goals, as with the management of the Columbia River (Thom
and O’Rourke 2005) or Laurentian Great Lakes (Bertram et
al. 1999; Neilson et al. 2003; Shear et al. 2003). The selection
of metrics is often made on the basis of specific management
objectives (Pantus and Dennison 2005) to target the effect of
specific stressors (Moss et al. 2003), or for cross-system com-
parative purposes (Dobiesz et al. 2010). The multiple-metric
approach is useful because it reflects multiple sources of stress-
ors (Karr and Chu 1999), although simplification to a single
number may mask important differences among sites (Norris
and Hawkins 2000).
Alternatively, ecosystem processes can be used as integra-
tive indices of ecosystem health. Ecosystem processes are often
driven by a wide range of physical, chemical, and biological
factors, making them natural integrators of system health. For
instance, leaf decomposition rates and organic matter reten-
tion have been used as an index of overall ecosystem health
(Wallace et al. 1996; Quinn et al. 2007), in which a higher
rate of decomposition and greater organic matter retention
is indicative of a healthier ecosystem. In addition, nitrifica-
tion (Hill et al. 2000) and nutrient uptake rates (Sabater et al.
2000) can be used as indicators of ecosystem health, with a
more retentive, tightly-coupled system considered healthier.
Perhaps the most fundamental of ecosystem processes, the
creation and consumption of organic matter by primary pro-
duction and respiration, are increasingly being used to assess
ecosystem activity, as well as ecosystem health. These rates are
often quantified using free-water measurements of dissolved
oxygen (Staehr et al. 2010).
Whereas oxygen concentration or biological oxygen
demand are established indicators of ecosystem health, espe-
cially with regard to wastewater discharge, only recently have
metabolism measurements become widespread enough to
be used as an indicator of ecosystem health. Measurements
of ecosystem metabolism have most commonly been used
to identify incidents of eutrophication in lakes, streams,
and coastal margins (Oviatt et al. 1986; Smith et al. 2005;
Matthews and Effler 2006; Kemp et al. 2009; Gucker et al.
2009). Metabolic rates also have been used to provide a holis-
tic measure of the effects of physical disturbances, such as
changes in flow regimes or turbidity in streams (Wiley et al.
1990; Floder and Sommer 1999; Young and Huryn 1999) or
flooding in lakes (Tsai et al. 2005; Sadro and Melack 2012).
Ecosystem metabolism has also been used as an indicator of
changes in catchment processes associated with agriculture
and industrial use (Wiley et al. 1990; Wilcock et al 1998;
Young and Huryn 1999; Sanders et al. 2007; Williamson et
al. 2008) or to demonstrate the effect of toxins or pollutants
(Giddings and Eddlemon 1978; Laursen et al. 2002; Wiegner et
al. 2003). Despite such widespread use, few studies use metab-
olism explicitly as a broad indicator of ecosystem health. This
is partly due to the challenge of interpreting metabolic rates as
indicators of ecosystem health, although some early attempts
suggest that it is possible (Young et al. 2008).
Section 9. Advantages and disadvantages of
integrative metrics
There are many advantages to using integrative metrics in
assessing ecosystem heath. By incorporating a broader diver-
sity of individual elements, they condense a wide variety of
environmental factors (Niemeijer 2002), attempt to account
for the complexity of aquatic systems and accommodate link-
ages between other components of the landscape. Ecosystem
process metrics, such as nutrient uptake rates or ecosystem
metabolism, provide a direct measure of ecosystem function.
Integrative metrics, by virtue of incorporating multiple ele-
ments, should be less sensitive to small-scale environmental
variability that complicates the interpretation of physical,
chemical, or biological metrics. Technological and method-
ological advances have made measurement of many of these
processes relatively straightforward. For example, ecosystem
metabolism can be easily measured through deployment of
automated sensors (Levi et al. 2013; Solomon et al. 2013). This
technology allows for the continuous collection of metabo-
lism-based metric data, increasing the temporal resolution of
the data, and eliminating the problem of missing important
ecosystem perturbation events or sampling during nonrepre-
sentative conditions. However, the ease of making measure-
ments does not remove the complications of interpreting such
data in the context of ecosystem health.
Despite these advantages, few integrative metrics have
received widespread use. Of those we have described, only tro-
phic state index (TSI) is used regularly to monitor ecosystem
health. It is an important component in water quality monitor-
ing among many organizations, from the USEPA to individual
lake or watershed associations (Carlson and Simpson 1996;
USEPA 2000a). In a comparative analysis involving more than
30 lakes, Jorgensen et al. (2005) demonstrated a strong linear
relationship between TSI and two more complex models of
ecosystem health, suggesting that it does a good job of char-
acterizing ecosystem health, despite its simplicity. Although
not as widely used, some ecosystem process metrics are prom-
ising, including ecosystem metabolism in rivers (Young et
al. 2008). The remaining metrics, which have primarily been
used only in academic studies, have failed to gain traction
largely because of the difficulty in translating measurements of
ecosystem process to an index that can be interpreted across
aquatic ecosystems and management schemes.
In addition, there are a number of complexities associated
Roley et al. Aquatic ecosystem health metrics
110
with the use of ecosystem process metrics. Two immediate
challenges are 1) identifying which physical, chemical, and
biological elements to include as proxies for ecosystem health
(Costanza et al. 1992; Patil et al. 2001; Schaeffer et al. 1988),
and 2) determining how to interpret them in a unified way.
As with other metrics, there remains the issue of interpreting
the condition of a specific site in the context of environmental
variability, as well as the selection of reference sites (Dobiesz
et al. 2010). Although ecosystem process-based metrics inher-
ently provide an integrative assessment of ecosystem health,
the myriad of factors that affect such processes, some of
which may operate in opposition, make interpreting them in
the context of ecosystem health difficult without a complete
understanding of the system (Young et al. 2008; Reuther 1992;
Wiegner et al. 2003). For example, low rates of gross primary
production could indicate impairment of biota or reflect a
system with naturally low rates due to climate and geology
(Young et al. 2008). Likewise, high daily variability in these
metrics makes long-term data sets and seasonal averaging
important when interpreting changes in system dynamics
(Dobiesz et al. 2010; Staehr et al. 2010; Coloso et al. 2011).
Despite the appeal of integrating across biological, chemical,
and physical ecosystem components, process metrics require a
greater expertise and understanding of an ecosystem, perhaps
limiting their application by small or volunteer-based moni-
toring programs.
Section 10. Advantages and challenges across
ecosystem health metrics
The decision to incorporate metrics into management is
influenced by the effort and expertise required to collect and
interpret metric data. Numerous metrics can be completed
rapidly and inexpensively by volunteers and citizen scientists,
which can increase the number of sites monitored and the
frequency of assessment, compared with metrics that require
expensive equipment and specialized personnel. In general,
though, rapid assessments provide less information than the
more sophisticated metrics. As a result, the goals of a mon-
itoring effort must be carefully considered before choosing
metrics. For example, simple visual assessment may be a useful
first step that identifies ecosystems for further study. Similarly,
sophisticated metrics might be reserved for sites with impend-
ing management decisions, those of high economic or eco-
logical importance, or sites that are particularly high-profile,
whereas the simpler metrics may be used more broadly to
quickly characterize an array of sites.
Many of the well-established metrics are based on a
comparison with reference conditions; thus, the choice of a
reference system is crucial to the validity of the results. They
rely on defining an undisturbed reference (or a more realistic
“Least-Disturbed Condition,” Stoddard et al. 2005) against
which to evaluate an ecosystem of interest. Identifying and
quantifying any reference system, let alone one relevant to
the ecosystem of interest, is a source of uncertainty because of
natural variation in ecosystem characteristics, such as physical
stability, biodiversity, chemical concentrations, and rates of
ecosystem processes. In addition, some systems may have no
known or measurable reference conditions. This limitation
has been addressed by metrics that combine theoretical and
empirical approaches (e.g., environmental flows, Poff et al.
1997), by metrics that are based on human health outcomes
(e.g., populations of harmful bacteria, USEPA 2010), and by
ecosystem classification modeling (e.g., grouping water bodies
by their relationship between land use and nutrient concen-
tration, Soranno et al. 2010). The continued development of
these approaches will be an important contribution to the
quantification of ecosystem health.
Finally, caution must be applied in the interpretation of any
individual metric. By necessity, each metric addresses a subset
of possible ecosystem health indicators, each subject to its own
uncertainties, biases, and assumptions. Ideally, before making
management decisions, managers will apply several metrics,
each addressing different aspects of ecosystem health. Such an
approach will provide a more complete picture of ecosystem
health, analogous to doctors using multiple tests to assess a
patient’s health (e.g., blood pressure alone is insufficient for
determining health). In addition, different metrics may pro-
vide complementary information; for example, a biological
assessment may reveal a preponderance of pollution-tolerant
taxa, whereas chemical or physical assessments may reveal
the cause of the biological impairment. As a result, it may
be useful to consider an individual metric as nested within
a larger health assessment. This approach has gained trac-
tion recently. For example, the USEPA’s National Wetlands
Condition Assessment (USEPA 2011b) and the National
Lakes Assessment (USEPA 2011a) both measure numerous
metrics of biological, physical, and chemical health. In doing
so, they are able to identify sites with impaired biotic commu-
nities, as well as the causes of that impairment.
Section 11. Gaps in existing metrics
There are currently numerous metrics available to assess
nearly all aspects of ecosystem health in aquatic systems. Some
metrics have a more formally developed framework of assess-
ment than others (e.g., biological versus integrative metrics),
and for a number of specific metrics we have identified places
where additional refinement of application and interpretation
would be useful and places where metrics are lacking, includ-
ing those that address the effects of emerging contaminants
and pharmaceuticals. In addition to these specific gaps, lim-
itations remain in the general use of ecosystem health metrics,
including a lack of uncertainty estimates, lack of inter-metric
comparisons, and lack of coordination of sampling efforts.
A lack of uncertainty estimates for most metrics fur-
ther complicates their interpretation; the assignment of a
single score for a given metric may be too parsimonious.
Despite a lack of replication, there are a number of alterna-
tive approaches that may be used to estimate uncertainty.
Roley et al. Aquatic ecosystem health metrics
111
Estimates could be calculated from the number of sampling
dates. For example, if a chemical concentration is measured
only once, the uncertainty associated with that value would be
high. Similarly, it could be calculated by comparing individual
components of an index, where individual components with
high uncertainties would result in an overall higher uncer-
tainty. Such an approach would report a range or distribu-
tion, rather than a single value. These uncertainty estimates
themselves might have important management implications,
where ecosystems with large uncertainties would possibly
receive further study before management decisions were
made. Interpreting uncertainty estimates along with metrics
of ecosystem health may help managers predict the likelihood
of improvement from intervention. However, communicating
uncertainty to the public remains complex. Whereas it is easy
to interpret a single number, the public is generally less famil-
iar with parameter distributions. Thus, if policy decisions are
based on such an approach, the results must be presented in a
way that makes sense to decision-makers and the public.
Whereas different metrics can provide similar information,
there may be large differences in the effort required. For exam-
ple, Wallace et al. (1996) demonstrated that leaf decomposition
rates provided similar results to the percent Ephemeroptera,
Plecoptera, and Trichoptera (%EPT), but % EPT was less
labor- and time-intensive. These comparative studies can help
ascertain which metrics overlap and which are most informa-
tive. Ultimately, a collection of these studies can be used to
determine the best way to allocate monitoring resources and
to determine where a particular metric is most appropriate.
Thus far, few such studies exist, but these comparisons may be
a fruitful area of future research. One particularly interesting
application would be to compare integrative metrics, such as
ecosystem metabolism or nutrient uptake, with easier-to-mea-
sure metrics such as nutrient concentration.
Finally, the vast number of metrics used across a myriad
of monitoring programs constrains large-scale analyses and
intercomparisons of metrics. To some degree, this is inevita-
ble, as some metrics may be more appropriate for particular
locations or ecosystem types. However, a standardized set of
protocols and a depot for data-sharing may result in analyses
that reveal broad spatial patterns. For example, the Global
Lakes Ecological Observatory Network (GLEON) was able to
identify the drivers of ecosystem respiration in lakes as a result
of data collected in a consistent, standardized manner through
a coordinated sensor network (Solomon et al. 2013).
Section 12. Priorities for future research
In addition to addressing the gaps in current metrics,
future research can use technological, computational, and
interdisciplinary tools to move the field forward. Ideally, these
efforts will move beyond existing frameworks to create new
paradigms for ecosystem health assessments, including the
development of early warning metrics, the explicit inclusion
of public input, and the use of complex systems models.
Improvements in these key areas may help improve ecosystem
health assessments and ultimately improve our ability to man-
age aquatic ecosystems.
Assessments of ecosystem health are often completed after
an environmental impact has occurred, but metrics that can
provide early warning of declines—analogous to preventative
medicine—are likely more effective at preventing damage to
ecosystems (Boulton et al. 1999). Early warning metrics may
be aided by the increasing availability of automated sensors,
and research priorities in this area should focus on identi-
fying the early warning signals. These warning signals may
be similar to those from the current suite of metrics, such as
critical levels of a contaminant or a low-oxygen threshold, but
the high frequency measurements allow the development of
novel metrics. For example, theoretical and experimental work
with long-term data sets suggest that the variance associated
with a specific metric, the return time of a metric to baseline
levels after a perturbation, or changes in the autocorrelation
of a metric’s time series may serve as early warnings of regime
shifts (Scheffer and Carpenter 2003; Carpenter et al. 2011;
Batt et al. 2013). These metrics use more complicated statistics
than most existing metrics, and thus they may require further
refinement before they are used broadly; water resource man-
agers may not have access to the necessary statistical tools and
policy-makers may not understand the terminology. These
problems are not intractable, however, and can likely be over-
come with careful consideration of these difficulties.
Automated sensors can also capture pulse events, such
as dips in dissolved oxygen, rapid changes in pH, or sudden
increases in sediment or nutrient concentrations. An increase
in frequency or duration of acute events may indicate prob-
lems before they become evident in biological surveys. This
approach will require research into baseline conditions, to
establish a healthy range and timing of pulse events. For
example, some sites may regularly experience periods of low
dissolved oxygen, during summer low-flow conditions, to
which the biota are well-adapted. Sensors do not yet exist for
all potential stressors, which limits the types of pulse events
that can be detected. Nonetheless, the current sensors provide
a good starting point for the development of metrics, and ini-
tial research efforts will help determine the utility of applying
this approach more broadly.
Early warning metrics will only be useful if they invoke
meaningful actions. In addition to knowing what an early
warning signal looks like, practitioners must have sufficient
system-specific knowledge to understand the causes of impair-
ment and know how to mitigate them. For all metrics, the
mitigation process may be smoother if citizen and stakeholder
input are explicitly included in a monitoring program. After
all, identifying a “healthy” or “unhealthy” system is only the
first step; mitigation requires the cooperation of numerous
stakeholders.
Inclusion of public input may improve the iterative process
between monitoring and management, and deserves further
Roley et al. Aquatic ecosystem health metrics
112
research. So far, public input is included in integrative models,
which include stakeholder preferences, the policy environ-
ment, and scientific knowledge (Croke et al. 2007). Similarly,
scenario-based models predict ecosystem responses to alterna-
tive management scenarios (e.g., Xu et al. 2013; Einheuser et
al. 2012). In these models, stakeholders can see how different
public policy options will influence water quality. Public input
could also be included in simpler ways, such as through a sur-
vey that establishes the expected usage of the system or accept-
able changes in land use or policy. Research in this area may
benefit from collaborations with sociologists and economists,
who can identify the relevant metrics for citizen input.
Physical, chemical, and biological ecosystem health metrics
are often presented as discrete ways of measuring and under-
standing ecosystem health—indeed, we make that distinction
in this chapter—but often, it is useful to consider multiple
types of metrics. Integrative metrics, as discussed above, are
one way to incorporate multiple ecosystem elements, but
another priority for future research is the coupling of com-
plex systems models. There are currently good models for
each aspect of ecosystem health, and combining them into a
larger, mechanistic model may improve our understanding
of the causes of impairment. These models are not likely to
be used as metrics themselves but instead will be applied to
impaired water bodies, with the intention of pinpointing the
sources of impairment, and ultimately improving manage-
ment. Development of these models will likely require the
collaboration of community ecologists, ecosystem ecologists,
geochemists, and computer scientists.
References
Aparicio, E., G. Carmona-Catot, P. B. Moyle, and E. Garcia-
Berthou. 2011. Development and evaluation of a fish-
based index to assess biological integrity of Mediterranean
streams. Aquat. Conserv. Mar. Freshw. Ecosys. 21:324-337
[doi:10.1002/aqc.1197].
Arthington, A. H., S. E. Bunn, N. L. Poff, and R. J. Naiman.
2006. The challenge of providing environmental flow
rules to sustain river ecosystems. Ecol. Appl. 16:1311-
1318 [doi:10.1890/1051-0761(2006)016[1311:TCOPE-
F]2.0.CO;2].
Attrill, M. J., and M. H. Depledge. 1997. Community and
population indicators of ecosystem health: Targeting links
between levels of biological organization. Aquat. Toxicol.
38:183-197 [doi:10.1016/S0166-445X(96)00839-9].
Batt, R. D., S. R. Carpenter, J. J. Cole, M. L. Pace, and R. A.
Johnson. 2013. Changes in ecosystem resilience detected
in automated measures of ecosystem metabolism during a
whole-lake manipulation. Proc. Nat. Acad. Sci. 110:17398-
17403 [doi:10.1073/pnas.1316721110].
Boulton, A.J. 1999. An overview of river health assessment:
philosophies, practice, problems and prognosis. 41: 469-
479 [doi:10.1046/j.1365-2427.1999.00443.x].
Bowman, M. F., and K. M. Somers. 2005. Considerations when
using the Reference Condition Approach for bioassess-
ment of freshwater ecosystems. Water Qual. Res. J. Can.
40:347-360
Beck, M. W., and L. K. Hatch. 2009. A review of research on
the development of lake indices of biotic integrity. Environ.
Rev. 17:21-44 [doi:10.1139/A09-001].
———, ———, B. Vondracek, and R. D. Valley. 2010.
Development of a macrophyte-based index of biotic
integrity for Minnesota lakes. Ecol. Indic. 10:968-979
[doi:10.1016/j.ecolind.2010.02.006].
Bertram, P., H. Shear, N. Stadler-Salt, and P. Horvatin. 1999.
Environmental and socioeconomic indicators of Great
Lakes basin health, pp. 703-720. In D. Rapport, W. Lasley,
D. Rolston, N. Nielsen, C. Qualset, and A. Damania (Eds.),
Managing for healthy ecosystems. Lewis Publishers.
Bogan, M. T., and K. S. Boersma. 2012. Aerial dispersal
of aquatic invertebrates along and away from arid-land
streams. Freshw. Sci. 31:1131-1144 [doi:10.1899/12-066.1].
Bovee, K. D., B. L. Lamb, J. M. Bartholow, C. B. Stalnaker,
J. Taylor, and J. Henriksen. 1998. Stream habitat analysis
using the instream flow incremental methodology. U.S.
Geological Survey Information and Technology Report
1998-0004. USGS.
Brown, B. L., and C. M. Swan. 2010. Dendritic net-
work structure constrains metacommunity proper-
ties in riverine ecosystems. J. Animal Ecol. 79:571-580
[doi:10.1111/j.1365-2656.2010.01668.x].
Bruhn, L. C., and P. A. Soranno. 2005. Long-term (1974-2001)
volunteer monitoring of water clarity trends in Michigan
lakes and their relation to ecoregion and land use/cover.
Lake Reservoir Manage. 21:10-23.
Buerge, I. J., T. Poiger, M. D. Muller, and H. R. Buser. 2003.
Caffeine, an anthropogenic marker for wastewater contam-
ination of surface waters. Environ. Sci. Technol. 37:691-700
[doi:10.1021/es020125z].
Bunn, S. E., and P. M. Davies. 2000. Biological processes
in running waters and their implications for the assess-
ment of ecological integrity. Hydrobiologia 422:61-70
[doi:10.1023/A:1017075528625].
Burton, T. M., D. G. Uzarski, J. P. Gathman, J. A. Genet, B. E.
Keas, and C. A. Stricker. 1999. Development of a prelimi-
nary invertebrate index of biotic integrity for Lake Huron
coastal wetlands. Wetlands 19:869-882 [doi:10.1007/
BF03161789].
Carlson, R. E. 1977. A trophic state index for lakes. Limnol.
Oceanogr. 22:361-369 [doi:10.4319/lo.1977.22.2.0361].
———, and J. Simpson. 1996. A coordinator’s guide to vol-
unteer lake monitoring methods. North American Lake
Management Society.
Carpenter, S. R., N. F. Caraco, D. L. Correll, R. W. Howarth,
A. N. Sharpley, and V. H. Smith. 1998. Nonpoint pollution
of surface waters with phosphorus and nitrogen. Ecol.
Appl. 8:559-568 [doi:10.1890/1051-0761(1998)008[0559:N-
POSWW]2.0.CO;2].
Roley et al. Aquatic ecosystem health metrics
113
———, and others. 2011. Early warnings of regime shifts:
A whole-ecosystem experiment. Science 332:1079-1082
[doi:10.1126/science.1203672].
Cleuvers, M. 2003. Aquatic ecotoxicity of pharmaceuticals
including the assessment of combination effects. Toxicol.
Lett. 142:185-194 [doi:10.1016/S0378-4274(03)00068-7].
Coloso, J. J, J. Cole, and M. Pace. 2011. Short-term varia-
tion in thermal stratification complicates estimation of
lake metabolism. Aquat. Sci. 73:305-315 [doi:10.1007/
s00027-010-0177-0].
Corsi, I., M. Mariottini, C. Sensini, L. Lancini, and S. Focardi.
2003. Fish as bioindicators of brackish ecosystem health:
integrating biomarker responses and target pollutant
concentrations. Oceanol. Acta 26:129-138 [doi:10.1016/
S0399-1784(02)01237-9].
Costanza, R., B. G. Norton, and B. D. Haskell (eds.). 1992.
Ecosystem health: new goals for environmental manage-
ment. Island Press.
Croke, B. F. W., J. L. Ticehurst, R. A. Letcher, J. P. Norton, L.
T. H. Newham, and A. D. Jakeman. 2007. Integrated assess-
ment of water resources: Australian experiences. Water
Res. Manage. 21:351-373 [doi:10.1007/s11269-006-9057-8].
Cummins, K. W., and M. J. Klug. 1979. Feeding ecology of
stream invertebrates. Ann. Rev. Ecol. System. 10:147-172
[doi:10.1146/annurev.es.10.110179.001051].
Cunningham, V. I. 2008. Special characteristics of pharma-
ceuticals related to environmental fate. pp 23-33. In K.
Kummerer (ed.), Pharmaceuticals in the environment.
Springer [doi:10.1007/978-3-540-74664-5_2].
Davies, P. E., J. H. Harris, T. J. Hillman, and K. F. Walker.
2010. The sustainable rivers audit: assessing river ecosys-
tem health in the Murray-Darling Basin, Australia. Mar.
Freshw. Res. 61:764-777 [doi:10.1071/MF09043].
Davis, W. S., B. D. Snyder, J. B. Stribling, and C. Stoughton.
1996. Summary of state biological assessment programs
for streams and wadeable rivers. EPA 230-R-96-007.
U.S. Environmental Protection Agency: Office of Policy,
Planning, and Evaluation.
Deblonde, T., C. Cossu-Leguille, and P. Hartemann. 2011.
Emerging pollutants in wastewater: A review of the litera-
ture. Int. J. Hyg. Environ. Health 214:442-448 [doi:10.1016/j.
ijheh.2011.08.002].
Diaz, R. J., and R. Rosenberg. 2008. Spreading dead zones and
consequences for marine ecosystems. Science 321:926-929
[doi:10.1126/science.1156401].
Dobiesz, N. E., and others. 2010. Metrics of ecosystem status
for large aquatic systems—A global comparison. J. Great
Lakes Res. 36:123-138 [doi:10.1016/j.jglr.2009.11.003].
Dodds, W. K., and R. M. Oakes. 2004. A technique for estab-
lishing reference nutrient concentrations across watersheds
affected by humans. Limnol. Oceanogr. Methods 2:333-341
[doi:10.4319/lom.2004.2.333].
Driscoll, C. T., G. B. Lawrence, A. J. Bulger, T. J. Butler, C. S.
Cronan, C. Eagar, and K. C. Weathers. 2001. Acidic deposition
in the northeastern United States: Sources and inputs,
ecosystem effects, and management strategies. Bioscience
51:180-198 [doi:10.1641/0006-3568(2001)051[0180:ADIT-
NU]2.0.CO;2].
Dyson, M., G. J. J. Bergkamp, and J. Scanlon (eds.). 2003. Flow:
The essentials of environmental flows. International Union
for Conservation of Nature and Natural Resources (IUCN)
[doi:10.2305/IUCN.CH.2003.WANI.2.en].
Einheuser, M. D., A. P. Nejadhashemi, S. P. Sowa, L. Wang,
Y. A. Hamaamin, and S. A. Woznicki. 2012. Modeling
the effects of conservation practices on stream health.
Sci. Total Environ. 435-436:380-391 [doi:10.1016/j.
scitotenv.2012.07.033].
Elser, J. J., M. E. S. Bracken, E. E. Cleland, D. S. Gruner, S. W.
Harpole, H. Hillebrand, and J. E. Smith. 2007. Global analysis
of nitrogen and phosphorus limitation of primary produc-
ers in freshwater, marine and terrestrial ecosystems. Ecol.
Lett. 10:1135-1142 [doi:10.1111/j.1461-0248.2007.01113.x].
Floder, S., and U. Sommer. 1999. Diversity in planktonic
communities: an experimental test of the intermediate
disturbance hypothesis. Limnol. Oceanogr. 44:1114-1119
[doi:10.4319/lo.1999.44.4.1114].
Giddings, J. M., and G. K. Eddlemon. 1978. Photosynthesis/
respiration ratios in aquatic microcosms under arsenic
stress. Water Air Soil Pollut. 9:207-212 [doi:10.1007/
BF00280707].
Gillette, D. P., J. S. Tiemann, D. R. Edds, and M. L.
Wildhaber, 2006. Habitat use by a midwestern USA
riverine fish assemblage: Effects of season, water tem-
perature and river discharge. J. Fish Biol. 68:1494-1512
[doi:10.1111/j.0022-1112.2006.001037.x].
Goldberg, C. S., D. S. Pilliod, R. S. Arkle, and L. P. Waits.
2011. Molecular detection of vertebrates in stream water:
A demonstration using Rocky Mountain Tailed Frogs and
Idaho Giant Salamanders. PLOS One 6:e22746 [doi:10.1371/
journal.pone.0022746].
Gorman, O. T., D. L. Yule, and J. D. Stockwell. 2012. Habitat
use by fishes of Lake Superior. II. Consequences of diel hab-
itat use for habitat linkages and habitat coupling in near-
shore and offshore waters. Aquat. Ecosys. Health Manage.
15:355-368.
Griffitt, R. J., J. Luo, J. Gao, J. C. Bonzongo, and D. S. Barber.
2008. Effect of particle composition and species on toxicity
of metallic nanoparticles in aquatic organisms. Environ.
Toxicol. Chem. 27:1972-1978 [doi:10.1897/08-002.1].
Gucker, B., I. G. Boechat, and A. Giani. 2009. Impacts of agri-
cultural land use on ecosystem structure and whole-stream
metabolism of tropical Cerrado streams. Freshw. Biolo.
54:2069-2085 [doi:10.1111/j.1365-2427.2008.02069.x].
Guzy, J. C., E. D. McCoy, A. C. Deyle, S. M. Gonzalez,
N. Halstead, and H. R. Mushinsky. 2012. Urbanization
interferes with the use of amphibians as indicators of
ecological integrity of wetlands. J. Appl. Ecol. 49:941-952
[doi:10.1111/j.1365-2664.2012.02172.x].
Roley et al. Aquatic ecosystem health metrics
114
Harper, D. M., J. L. Kemp, B. Vogel, and M. D. Newson.
2000. Towards the assessment of ‘ecological integrity’ in
running waters of the United Kingdom. Hydrobiologia
422/423:133-142 [doi:10.1023/A:1017072906760].
Hauer, F. R. and G. A. Lamberti (eds.). 2007. Methods in
stream ecology. Elsevier.
Herlihy, A. T., N. C. Kamman, J. C. Sifneos, D. Charles,
M. D. Enache, and R. J. Stevenson. 2013. Using mul-
tiple approaches to develop nutrient criteria for lakes
in the conterminous USA. Freshw. Sci. 32(2):67-384
[doi:10.1899/11-097.1].
Hill, B. H., R. K. Hall, P. Husby, A. T. Herlihy, and M.
Dunne. 2000. Interregional comparisons of sediment
microbial respiration in streams. Freshw. Biol. 44:213–222
[doi:10.1046/j.1365-2427.2000.00555.x].
Holtgrieve, G. W., and others. 2011. A coherent signature
of anthropogenic nitrogen deposition to remote water-
sheds of the Northern hemisphere. Science 334:1545-1548
[doi:10.1126/science.1212267].
Howarth, R. W., and R. Marino. 2006. Nitrogen as the limiting
nutrient for eutrophication in coastal marine ecosystems:
Evolving views over three decades. Limnol. Oceanogr.
51:364-376 [doi:10.4319/lo.2006.51.1_part_2.0364].
Jerde, C. L., A. R. Mahon, W. L. Chadderton, and D. M.
Lodge. 2011.”Sight-unseen” detection of rare aquatic spe-
cies using environmental DNA. Conserv. Lett. 4:150-157
[doi:10.1111/j.1755-263X.2010.00158.x].
Jorgensen, S. E., R. Costanza, and F. Xu. 2005. Handbook of eco-
logical indicators for assessment of ecosystem health. Taylor
and Francis, CRC Press [doi:10.1201/9780203490181].
Kane, D. D., S. I. Gordon, M. Munawar, M. N. Charlton,
and D. A. Culver. 2009. The Planktonic Index of Biotic
Integrity (P-IBI): An approach for assessing lake eco-
system health. Ecol. Indic. 9:1234-1247 [doi:10.1016/j.
ecolind.2009.03.014].
Karr, J. R. 1981. Assessment of biotic integrity using fish com-
munities. Fisheries 6:21-27 [doi:10.1577/1548-8446(1981)0
06<0021:AOBIUF>2.0.CO;2].
———, K. D. Fausch, P. L. Angermeier, P. R. Yant and I. J.
Schlosser. 1986. Assessing biological integrity in running
waters. A method and its rationale. Illinois Natural History
Survey. Special Publication No. 5.
———, and E. W. Chu. 1999. Restoring life in running
waters: Better biological monitoring. Island Press
[doi:10.1023/A:1005167831280].
Kaushal, S. S., G. E. Likens, R. Utz, M. L. Pace, M. Grese,
and M. Yepsen. 2013. Increased river alkalinization in the
Eastern U.S. Environ. Sci. Technol [doi:10.1021/es401046s].
Kemp, W. M., J. M. Testa, D. J. Conley, D. Gilbert, and J.
D. Hagy. 2009. Temporal responses of coastal hypoxia to
nutrient loading and physical controls. Biogeosciences
6:2985-3008 [doi:10.5194/bg-6-2985-2009].
Kerans, B. L., and J. R. Karr. 1994. A benthic index of biotic
integrity (B-IBI) for rivers of the Tennessee Valley. Ecol.
Appl. 4:768-785 [doi:10.2307/1942007].
Kolpin, D. W., E. T. Furlong, M. T. Meyer, E. M. Thurman,
S. D. Zaugg, L. B. Barber, and H. T. Buxton. 2002.
Pharmaceuticals, hormones, and other organic wastewa-
ter contaminants in US streams, 1999-2000: A national
reconnaissance. Environ. Sci. Technol. 36:1202-1211
[doi:10.1021/es011055j].
Kroll, A. J., M. P. Hayes, and J. G. MacCracken. 2009.
Concerns regarding the use of amphibians as met-
rics of critical biological thresholds: a comment on
Welsh & Hodgson (2008). Freshw. Biol. 54:2364-2373
[doi:10.1111/j.1365-2427.2009.02245.x].
Lackey, R. T. 2001. Values, policy, and ecosystem health.
Bioscience 51:437-443 [doi:10.1641/0006-3568(2001)051
[0437:VPAEH]2.0.CO;2].
Laursen, A. E., S. P. Seitzinger, R. Dekorsey, J. G. Sanders, D.
L. Breitburg, and R. W. Osman. 2002. Multiple stressors in
an estuarine system: effects of nutrients, trace elements, and
trophic complexity on benthic photosynthesis and respira-
tion. Estuaries 25:57-69 [doi:10.1007/BF02696049].
Lawrence, J. R., B. Zhu, G. D. W. Swerhone, J. Roy, V. Tumber,
M. J. Waiser, and D. R. Korber. 2012. Molecular and micro-
scopic assessment of the effects of caffeine, acetaminophen,
diclofenac, and their mixtures on river biofilm commu-
nities. Environ. Toxicol. Chem. 31:508-517 [doi:10.1002/
etc.1723].
Leibold, M. A., and others. 2004. The metacommunity con-
cept: a framework for multi-scale community ecology. Ecol.
Lett. 7:601-613 [doi:10.1111/j.1461-0248.2004.00608.x].
Levi, P. S., J. L. Tank, J. Ruegg, D. J. Janetski, S. D. Tiegs, D.
T. Chaloner, and G. A. Lamberti. 2013. Whole-stream
metabolism responds to spawning Pacific Salmon in their
native and introduced ranges. Ecosystems 16:269-283
[doi:10.1007/s10021-012-9613-4].
Likens, G. E., C. T. Driscoll, and D. C. Buso. 1996. Long-
term effects of acid rain: Response and recovery of a
forest ecosystem. Science 272:244-246 [doi:10.1126/
science.272.5259.244].
Maddock, I. 1999. The importance of physical habitat assess-
ment for evaluating river health. Freshw. Biol. 41:373-391
[doi:10.1046/j.1365-2427.1999.00437.x].
Mast, M. A., W. T. Foreman, and S. V. Skaates. 2007.
Current-use pesticides and organochlorine compounds
in precipitation and lake sediment from two high-eleva-
tion national parks in the Western United States. Arch.
Environ. Contamin. Toxicol. 52(3):294-305 [doi:10.1007/
s00244-006-0096-1].
Matthews, D. A., and S. W. Effler. 2006. Assessment of long-
term trends in the oxygen resources of a recovering urban
lake, Onondaga Lake, NY. Lake Reserv. Manage. 22:19-32
[doi:10.1080/07438140609353881].
McCormick, F. H., B. H. Hill, L. P. Parrish, and W. T.
Willingham. 1994. Mining impacts on fish assemblages in
the Eagle and Arkansas Rivers, Colorado. J. Freshw. Ecol.
Roley et al. Aquatic ecosystem health metrics
115
9:175-179 [doi:10.1080/02705060.1994.9664884].
McGoff, E., and others. 2013. Assessing the relationship
between the Lake Habitat Survey and littoral macroinverte-
brate communities in European lakes. Ecol. Indic. 24:205-
214 [doi:10.1016/j.ecolind.2012.09.018].
Meyer, J. L. 1997. Stream health: incorporating the human
dimension to advance stream ecology. J. N. Am. Benthol.
Soc. 16:439-447 [doi:10.2307/1468029].
Milhous, R. T., and T. J. Waddle. 2012. Physical Habitat
Simulation (PHABSIM) Software for Windows (v.1.5.1).
USGS Fort Collins Science Center.
Milly, P. C. D., K. A. Dunne, and A. V. Vecchia. 2005. Global
pattern of trends in streamflow and water availability
in a changing climate. Nature 438:347-350 [doi:10.1038/
nature04312].
Moss, B., and other. 2003. The determination of ecological qual-
ity in shallow lakes: a tested expert system (ECOFRAME)
for implementation of the European Water Framework
Directive. Aquat. Conserv. Mar. and Freshw. Syst. 13:507-
550 [doi:10.1002/aqc.592].
Murray, K. E., S. M. Thomas, and A. A. Bodour. 2010.
Prioritizing research for trace pollutants and emerging con-
taminants in the freshwater environment. Environ. Pollut.
158:3462-3471 [doi:10.1016/j.envpol.2010.08.009].
Nakada, N., K. Kiri, H. Shinohara, A. Harada, K. Kuroda, S.
Takizawa, and H. Takada. 2008. Evaluation of pharmaceu-
ticals and personal care products as water-soluble molecu-
lar markers of sewage. Environ. Sci. Technol. 42:6347-6353
[doi:10.1021/es7030856].
Neilson, M. A., and others. 2003. Ecological monitoring
for assessing the state of the nearshore and open waters
of the Great Lakes. Environ. Monit. Assess. 88:103-117
[doi:10.1023/A:1025500619900].
Niemeijer, D. 2002. Developing indicators for environmental
policy: data-driven and theory-driven approaches exam-
ined by example. Environ. Sci. Policy 5:91-103 [doi:10.1016/
S1462-9011(02)00026-6].
Niyogi, D. K., W. M. Lewis, and D. M. McKnight. 2002.
Effects of stress from mine drainage on diversity, biomass,
and function of primary producers in mountain streams.
Ecosystems 5:554-567 [doi:10.1007/s10021-002-0182-9].
Norris, R. H., and C. P. Hawkins. 2000. Monitoring river health.
Hydrobiologia 435:5-17 [doi:10.1023/A:1004176507184].
Oviatt, C. A., A. A. Keller, P. A. Sampou, and L. L. Beatty.
1986. Patterns of productivity during eutrophication: a
mesocosm experiment. Mar. Ecol. Progr. Ser. 28:69-80
[doi:10.3354/meps028069].
Palm, H. W., and S. Ruckert. 2009. A new approach to visu-
alize ecosystem health by using parasites. Parasitol. Res.
105:539-553 [doi:10.1007/s00436-009-1423-z].
Pantus, F. J., and W. C. Dennison. 2005. Quantifying and
evaluating ecosystem health: a case study from Moreton
Bay. Aust. Environ. Manage. 36:757-771 [doi:10.1007/
s00267-003-0110-6].
Parasiewicz, P. 2008. Habitat time-series analysis to define
flow-augmentation strategy for the Quinebaug River,
Connecticut and Massachusetts, USA. River Res. Appl.
24:439-452 [doi:10.1002/rra.1066].
———, and M. J. Dunbar. 2001. Physical habitat modelling
for fish—a developing approach. Arch. Hydrobiol. Supp.
135/2-4:1-30.
Patil, G. P., R. P. Brooks, W. L. Myers, D. J. Rapport,
and C. Taillie. 2001. Ecosystem health and its measure-
ment at landscape scale: toward the next generation
of quantitative assessments. Ecosyst. Health 7:307-316
[doi:10.1046/j.1526-0992.2001.01034.x].
Patrick, C. J., and C. M. Swan. 2011. Reconstructing the
assembly of a stream-insect metacommunity. J. N. Am.
Benthol. Soc. 30:259-272 [doi:10.1899/09-169.1].
Petrone, K. C., J. B. Jones, L. D. Hinzman, and R. D.
Boone. 2006. Seasonal export of carbon, nitrogen, and
major solutes from Alaskan catchments with discontin-
uous permafrost. J. Geophys. Res. Biogeosci. 111(G2)
[doi:10.1029/2005JG000055].
Pfankuch, D. J. 1975. Stream reach inventory and channel sta-
bility evaluation. USDA Forest Service. Washington, D.C.
RI-75-002.
Pietrock, M., and O. Hursky. 2011. Fish and ecosystem health
as determined by parasite communities of lake white-
fish (Coregonus clupeaformis) from Saskatchewan boreal
lakes. Water Qual. Res. J. Can. 46:219-229 [doi:10.2166/
wqrjc.2011.004].
Pilliod, D. S., C. S. Goldberg, R. S. Arkle, and L. P. Waits. 2013.
Estimating occupancy and abundance of stream amphib-
ians using environmental DNA from filtered water sam-
ples. Can. J. Fish. Aquat. Sci. 70:1123-1130 [doi:10.1139/
cjfas-2013-0047].
Poff, N. L., and others. 1997. The natural flow regime: a
new paradigm for riverine conservation and restoration.
BioScience 47:769-784 [doi:10.2307/1313099].
———, J. D. Olden, N. K. M. Vieira, D. S. Finn, M. P.
Simmons, and B. C. Kondratieff. 2006. Functional trait
niches of North American lotic insects: traits-based ecolog-
ical applications in light of phylogenetic relationships. J. N.
Am. Benthol. Soc. 25:730-755 [doi:10.1899/0887-3593(200
6)025[0730:FTNONA]2.0.CO;2].
Pont, D., R. M. Hughes, T. R. Whittier, and S. Schmutz. 2009.
A predictive index of biotic integrity model for aquatic-ver-
tebrate assemblages of Western US streams. Trans. Am.
Fish. Soc. 138:292-305 [doi:10.1577/T07-277.1].
Quinn, J. M., N. R. Phillips, and S. M. Parkyn. 2007. Factors
influencing retention of coarse particulate organic matter
in streams. Earth Surf. Process. Landforms 32:1186-1203
[doi:10.1002/esp.1547].
Rankin, E. T. 1989. The qualitative habitat evaluation index
(QHEI): rationale, methods, and application. State of Ohio,
Environmental Protection Agency, Division of Water
Quality. Columbus, Ohio.
Roley et al. Aquatic ecosystem health metrics
116
Regier, H. A. 1993.The notion of natural and cultural integrity,
pp. 3-18. In S. J. Woodley, J. J. Kay, and G. Francis (eds.),
Ecological integrity and the management of ecosystems. St.
Lucie Press.
Reuther, R. 1992. Arsenic introduced into a littoral fresh-
water model ecosystem. Sci. Total Environ. 115:219-237
[doi:10.1016/0048-9697(92)90331-L].
Roulet, N., and T. R. Moore. 2006. Environmental chem-
istry: Browning the waters. Nature 444:283-284
[doi:10.1038/444283a].
Ruaro, R., and E. A. Gubiani. 2013. A scientometric assess-
ment of 30 years of the Index of Biotic Integrity in aquatic
ecosystems: Applications and main flaws. Ecol. Indic.
29:105-110 [doi:10.1016/j.ecolind.2012.12.016].
Sabater, F., A. Butturini, E. Marti, I. Munoz, A. Romani, J.
Wray, and S. Sabater. 2000. Effects of riparian vegetation
removal on nutrient retention in a Mediterranean stream.
J. N. Am. Benthol. Soc. 19:609-620 [doi:10.2307/1468120].
Sadro, S., and J. M. Melack. 2012. The effect of an extreme rain
event on the biogeochemistry and ecosystem metabolism of
an oligotrophic high-elevation lake. Arctic Antarc. Alpine
Res. 44:222-231 [doi:10.1657/1938-4246-44.2.222].
Sanders, I. A., C. M. Heppell, J. A. Cotton, G. Wharton, A. G.
Hildrew, E. J. Flowers, and M. Trimmer. 2007. Emission
of methane from chalk streams has potential implica-
tions for agricultural practices. Freshw. Biol. 52:1176-1186
[doi:10.1111/j.1365-2427.2007.01745.x].
Sanderson, H., D. J. Johnson, T. Reitsma, R. A. Brain, C. J.
Wilson, and K. R. Solomon. 2004. Ranking and prior-
itization of environmental risks of pharmaceuticals in
surface waters. Regulat. Toxicol. Pharmacol. 39:158-183
[doi:10.1016/j.yrtph.2003.12.006].
Schaeffer, D. J., E. E. Henricks, and H. W. Kerster. 1988.
Ecosystem health: 1. Measuring ecosystem health. Environ.
Manage. 12:445–455 [doi:10.1007/BF01873258].
Scheffer, M., and S. R. Carpenter. 2003. Catastrophic regime
shifts in ecosystems: linking theory to observation. Trends
Ecol. Evol. 18:648-656 [doi:10.1016/j.tree.2003.09.002].
Schindler, D. W. 1977. Evolution of phosphorus limita-
tion in lakes. Science 195:260-262 [doi:10.1126/
science.195.4275.260].
———, K. H. Mills, D. F. Malley, D. L. Findlay, J. A. Shearer,
I. J. Davies, and D. R. Cruikshank. 1985. Long-term ecosys-
tem stress—the effects of years of experimental acidifica-
tion on a small lake. Science 228:1395-1401 [doi:10.1126/
science.228.4706.1395].
Schubert, S., A. Peter, R. Burki, R. Schoenenberger, M.
J. F. Suter, H. Segner, and P. Burkhardt-Holm. 2008.
Sensitivity of brown trout reproduction to long-term estro-
genic exposure. Aquat. Toxicol. 90:65-72 [doi:10.1016/j.
aquatox.2008.08.002].
Schwab, B. W., E. P. Hayes, J. M. Fiori, F. J. Mastrocco, N.
M. Roden, D. Cragin,.and P. D. Anderson. 2005. Human
pharmaceuticals in US surface waters: A human health
risk assessment. Regulat. Toxicol. Pharmacol. 42:296-312
[doi:10.1016/j.yrtph.2005.05.005].
Schwarzenbach, R. P., B. I. Escher, K. Fenner, T. B. Hofstetter,
C. A. Johnson, U. von Gunten, and B. Wehrli. 2006. The
challenge of micropollutants in aquatic systems. Science
313:1072-1077 [doi:10.1126/science.1127291].
Shear, H., N. Stadler-Salt, P. Bertram, and P. Horvatin. 2003.
The development and implementation of indicators of
ecosystem health in the Great Lakes basin. Environ. Monit.
Assess. 88:119-152 [doi:10.1023/A:1025504704879].
Smith, S. V., R. W. Buddemeier, F. Wulff, and D. P. Swaney.
2005. C, N, P fluxes in the coastal zone, pp. 95-143. In C. J.
Crossland, H. H. Kremer, H. J. Lindeboom, J. I. Marshall-
Crossland, and M. D. A. Le Tissier (eds.), Coastal fluxes in
the anthropocene. Springer [doi:10.1007/3-540-27851-6_3].
Smith, V. H., G. D. Tilman, and J. C. Nekola. 1999.
Eutrophication: impacts of excess nutrient inputs on fresh-
water, marine, and terrestrial ecosystems. Environ. Pollut.
100(1-3):179-196 [doi:10.1016/S0269-7491(99)00091-3].
Solomon, C. T., and others. 2013. Ecosystem respiration:
Drivers of daily variability and background respiration
in lakes around the globe. Limnol. Oceanogr. 58:849-866
[doi:10.4319/lo.2013.58.3.0849].
Soranno, P. A., K. S. Cheruvelil, K. E. Webster, M. T. Bremigan,
T. Wagner, and C. A. Stow. 2010. Using landscape limnol-
ogy to classify freshwater ecosystems for multi-ecosys-
tem management and conservation. Bioscience 60:440-454
[doi:10.1525/bio.2010.60.6.8].
Soto-Galera, E., J. Paulo-Maya, E. Lopez-Lopez, J. A. Serna-
Hernandez, and J. Lyons. 1999. Change in fish fauna as
indication of aquatic ecosystem condition in Rio Grande de
Morelia-Lago de Cuitzeo Basin, Mexico. Environ. Manage.
24:133-140 [doi:10.1007/s002679900221].
Staehr, P. A., and others. 2010. Lake metabolism and the diel
oxygen technique: State of the science. Limnol. Oceanogr.
Methods 8:628-644 [doi:10.4319/lom.2010.8.0628].
Stoddard, J. L., D. S. Jeffries, A. Lukewille, T. A. Clair, P. J. Dillon,
C. T. Driscoll, and A. Wilander. 1999. Regional trends in
aquatic recovery from acidification in North America and
Europe. Nature 401:575-578 [doi:10.1038/44114].
———, and others. 2005. Environmental Monitoring and
Assessment Program (EMAP): Western streams and rivers
statistical summary. EPA 620/R-05/006, U.S. Environmental
Protection Agency.
———, A. T. Herlihy, D. V. Peck, R. M. Hughes, T. R. Whittier,
and E. Tarquinio. 2008. A process for creating multimetric
indices for large-scale aquatic surveys. J. N. Am. Benthol.
Soc. 27:878-891 [doi:10.1899/08-053.1].
Strange, E. M., K. D. Fausch, and A. P. Covich, 1999. Sustaining
ecosystem services in human- dominated watersheds:
Biohydrology and ecosystem processes in the South Platte
River Basin. Environ. Manage. 24(1):39-54 [doi:10.1007/
s002679900213].
Thom, R. M., and L. K. O’Rourke. 2005. Ecosystem health
Roley et al. Aquatic ecosystem health metrics
117
indicator metrics for the lower Columbia River and estuary
partnership. Battelle Marine Sciences Laboratory.
Thomsen, P. E., J. Kielgast, L. L. Iversen, C. Wiuf, M.
Rasmussen, M. T. P. Gilbert, L. Orlando, and E. Willerslev.
2012. Monitoring endangered freshwater biodiversity
using environmental DNA. Mol. Ecol. 21:2565-2573
[doi:10.1111/j.1365-294X.2011.05418.x].
Thornton, T., and J. Leahy. 2012. Trust in citizen science research:
A case study of the groundwater education through water
evaluation and testing program. J. Am. Water Res. Assoc.
48:1032-1040 [doi:10.1111/j.1752-1688.2012.00670.x].
Tsai, A. Y., K. P. Chiang, J. Chang, and G. C. Gong. 2005.
Seasonal diel variations of picoplankton and nanoplankton
in a subtropical western Pacific coastal ecosystem. Limnol.
Oceanogr. 50:1221-1231 [doi:10.4319/lo.2005.50.4.1221].
[USEPA] U.S. Environmental Protection Agency. 1989. Risk
assessment guidance for Superfund Volume I: human health
evaluation manual (part A). EPA 540-1-89-002. Office of
Emergency and Remedial Response, US Environmental
Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 1990.
Freshwater macroinvertebrate species list including tol-
erance values and functional feeding group designations
for use in rapid bioassessment protocols. EA Report
No.11075.05. U.S. Environmental Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 1991.
Technical support document for water quality based
toxics control. EPA 505-2-90-001. Office of Water, US
Environmental Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 1998.
National strategy for the development of regional nutri-
ent criteria. EPA-8221-R-98-002. Office of Water, US
Environmental Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 2000a.
Nutrient criteria technical guidance manual: lakes and
reservoirs. EPA-822-B00-001. Office of Water, US
Environmental Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 2000b.
Ambient water quality criteria recommendations.
Information supporting the development of state and
tribal nutrient criteria for lakes and reservoirs in nutrient
ecoregion VI. EPA 822-B-00-008. Office of Water, US
Environmental Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 2005.
National acid precipitation assessment program report to
Congress: an integrated assessment. US Environmental
Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 2010.
National lakes assessment: Technical appendix data anal-
ysis approach. EPA 841-R009-001a. Office of Water/Office
of Research and Development.
[USEPA] U.S. Environmental Protection Agency. 2011a. 2012
National Lakes Assessment. Field Operations Manual.
Office of Water. EPA 841-B-11-003. U.S. Environmental
Protection Agency.
[USEPA] U.S. Environmental Protection Agency. 2011b.
National Wetland Condition Assessment. Field Operations
Manual. EPA 843-R-10-001.
[USEPA] U.S. Environmental Protection Agency. 2012. Water
quality assessment and total maximum daily loads informa-
tion (http://www.epa.gov/waters/ir/). Office of Water, US
Environmental Protection Agency. <Web site accessed 25
June 2013>.
[USEPA]U.S. Environmental Protection Agency. 2013.
National rivers and streams assessment 2008-2009 tech-
nical report draft. U.S. Environmental Protection Agency
Office of Wetlands, Oceans, and Watersheds, Office of
Research and Development.
[VANR] Vermont Agency of Natural Resources. 2008.
Reach habitat assessments. Department of Environmental
Conservation and Fish and Wildlife Department.
Vitousek, P. M., J. D. Aber, R. W. Howarth, G. E. Likens, P.
A. Matson, D. W. Schindler, and D. Tilman. 1997. Human
alteration of the global nitrogen cycle: Sources and conse-
quences. Ecol. Appl. 7:737-750 [doi:10.2307/2269431].
Vugteveen, P., R. Leuven, M. A. J. Huijbregts, and H. J. R.
Lenders. 2006. Redefinition and elaboration of river ecosys-
tem health: perspective for river management. Hydrobiologia
565:289-308 [doi:10.1007/s10750-005-1920-8].
Wade, T. J., R. L. Calderon, E. Sams, M. Beach, K. P. Brenner,
A. H. Williams, and A. P. Dufour. 2006. Rapidly measured
indicators of recreational water quality are predictive of
swimming-associated gastrointestional illness. Environ.
Health Perspect. 114:24-28 [doi:10.1289/ehp.8273].
Wallace, J. B., J. W. Grubaugh, and M. R. Whiles. 1996.
Biotic indices and stream ecosystem processes: results
from an experimental study. Ecol. Appl. 6:140-151
[doi:10.2307/2269560].
Wang, L., J. Lyons, and P. Kanehl. 2003. Impacts of urban land
cover on trout streams in Wisconsin and Minnesota. Trans.
Am. Fish. Soc. 132:825-839 [doi:10.1577/T02-099].
Welsh, H. H., and G. R. Hodgson. 2008. Amphibians as
metrics of critical biological thresholds in forested head-
water streams of the Pacific Northwest, USA. Freshw. Biol.
53:1470-1488 [doi:10.1111/j.1365-2427.2008.01963.x].
Welter, J. R., S.G. Fisher, and N.B. Grimm. 2005. Nitrogen
transport and retention in an arid land watershed:
Influence of storm characteristics on terrestrial-aquatic
linkages. Biogeochemistry 76(3):421-440 [doi:10.1007/
s10533-005-6997-7].
Wetzel, R. G. 2001. Limnology: lake and river ecosystems. 3rd
edition. Academic Press.
Wheaton, J., J. Brasington, S. E. Darby, A. Kasprak, D. Sear,
and D. Vericat. 2013. Morphodynamic signatures of braid-
ing mechanisms as expressed through change in sediment
storage in a gravel-bed river. J. Geophys. Res. Earth Surf.
118 [doi:10.1002/jgrf.20060].
Wheaton, J. M., C. Gibbins, J. Wainwright, L. Larsen, and B.
Roley et al. Aquatic ecosystem health metrics
118
McElroy. 2011. Preface: Multiscale feedbacks in ecogeo-
morphology. Geomorphology 126:265-268 [doi:10.1016/j.
geomorph.2011.01.002].
Wiegner, T. N., S. P. Seitzinger, D. L. Breitburg, and J. G.
Sanders. 2003. The effects of multiple stressors on the
balance between autotrophic and heterotrophic processes
in an estuarine system. Estuaries 26:352-364 [doi:10.1007/
BF02695973].
Williamson, C. E., W. Dodds, T. K. Kratz, and M. A. Palmer.
2008. Lakes and streams as sentinels of environmental
change in terrestrial and atmospheric processes. Front.
Ecol. Environ. 6:247-254 [doi:10.1890/070140].
———, and others. 2014. Lakes as sensors in the land-
scape: Optical metrics as scalable sentinel responses to cli-
mate change. Limnol. Oceanogr. 59:840-850 [doi:10.4319/
lo.2014.59.3.0840].
Wilcock, R. J., J. W. Nagels, G. B. McBride, K. J. Collier, B. T.
Wilson, and B. A. Huser. 1998. Characterisation of lowland
streams using a single-station diurnal curve analysis model
with continuous monitoring data for dissolved oxygen and
temperature. New Zeal. J. Mar. Freshw. Res. 32:67-79 [doi:
10.1080/00288330.1998.9516806].
Wiley, M. J., L. L. Osbourne, and R. W. Larimore. 1990.
Longitudinal structure of an agricultural prairie river
system and its relationship to current stream ecosystem
theory. Can. J. Fish. Aquat. Sci. 47:373-384 [doi:10.1139/
f90-039].
Xu, F., Z. F. Yang, B. Chen, and Y. W. Zhao. 2013. Impact of
submerged plants on ecosystem health of the plant-domi-
nated Baiyangdian Lake, China. Ecol. Model. 252:167-175
[doi:10.1016/j.ecolmodel.2012.07.013].
Young, R. G., and A. D. Huryn. 1999. Effects of land use on
stream metabolism and organic matter turnover. Ecol.
Appl. 9:1359-1376 [doi:10.1890/1051-0761(1999)009[135
9:EOLUOS]2.0.CO;2].
———, C. D. Matthaei, and C. R. Townsend. 2008. Organic
matter breakdown and ecosystem metabolism: functional
indicators for assessing ecosystem health. J. N. Am. Benthol.
Soc. 27:605-625 [doi:10.1899/07-121.1].
... They also found that the limit values specified can provide biotic potentials that can be modified by other factors. Given that different environmental stressors affect the structure and functioning of the ecosystems to various degrees, integrative and multi-metric approaches give a more holistic representation of the state of the ecosystem, combining physical, chemical and biological measurements in one synthetic index (Revenga et al., 2005;Roley et al., 2014). Hirabayashi et al. (2015) studied the bathymetric distribution of benthic macroinvertebrates in a deep oligotrophic lake in Japan (Lake Table 2 Littoral biogenic index (Bl) (quantitative), taxonomic deficit (Df) and LBI during the monitoring stations used to calculate the LBI in Lake Rupanco during autumn and spring (year 2013). ...
Article
Full-text available
Increased pollution and degradation of water resources and their associated ecosystems has stimulated the development of tools and methodologies to characterize, estimate, predict, and reverse the environmental impact of anthropic effects on water bodies. The Secondary Water Quality Standards (NSCA) adopted in Chile have incorporated the use of bioindicators complementary to physicochemical analyses, in order to determine the ecological condition of lotic and lentic environments. Our research used the "Lake Biotic Index" (LBI) to establish the ecological condition of Lake Rupanco using benthic macroinvertebrates. The results indicated an Oligo-Eubiotic condition for this lake given the high concentration of oxygen and low organic matter content in sediments, in addition to low biogenic potential and good taxa preservation in both the autumn and spring surveys. Features of the ecological condition obtained through the application of the LBI (benthic subsystem) conform to the results of physicochemical and microalgae analyses undertaken previously in Lake Rupanco (pelagic subsystem). Based on these results, we support application of the LBI index as a complementary tool for the integrated management of lentic ecosystems.
... The timing of both DOC quantity and molecular characteristics impacts DOC bioavailability and, therefore, timing of DOC flux can have a large effect on organisms, downstream ecosystems, and water quality. These in-situ sensors seem to be capable of greatly improving our ability to "take the pulse" of key ecosystem parameters, specifically, DOC moving through watersheds and ecosystems (Roley et al., 2014). In taking the pulse of DOC in streams, we are also able to use that information to guide the study design and monitoring of other watershed attributes that control ecosystems and water quality, such as limiting nutrients (Blaen et al., 2016). ...
Article
It is important to understand how dissolved organic carbon (DOC) is processed and transported through stream networks because DOC is a master water quality variable in aquatic ecosystems. High-frequency sampling is necessary to capture important, rapid shifts in DOC source, concentration, and composition (i.e. quality) in streams. Until recently, this high-frequency sampling was logistically difficult or impossible. However, this type of sampling can now be conducted using in-situ optical measurements through long-term, field-deployable fluorometers and spectrophotometers. The optical data collected from these instruments can quantify both DOC concentration and composition properties (e.g., specific ultra-violet absorbance at 254 nm, spectral slope ratio, and fluorescence index). Previously, the use of these sensors was limited to a small number of specialized users, mainly in Europe and North America, where they were used predominantly in marine DOC studies as well as water treatment and management infrastructure. However, recent field demonstrations across a wide range of river systems reveals a large potential for the use of these instruments in freshwater environments, heightening interest and demand across multiple environmental research and management disciplines. Hence, this review provides an up-to-date synthesis on 1) the use of spectroscopy as a diagnostic tool in stream DOC studies, 2) the instrumentation, its applications, potential limitations and future considerations, and 3) the new watershed DOC research directions made possible via these in-situ optical sensors.
... As a consequence, water supply reservoir catchments in these uplands are generally managed (stabilization of soils, moorland restoration and protection of watercourses) to ensure the provision of good water quality (United Utilities, 2011), with major focus on DOC removal; the most significant cost to water treatment in the UK (Worrall et al., 2004). Whilst soil erosion and sediment transport of particulate nutrients to reservoirs and impoundments constitute redistribution or translocation of nutrient sinks within catchments (Ittekkot and Zhang, 1989;Downing et al., 2008;Battin et al., 2009), the focus of our study was on the biologically active component of Ndissolved N; one of the most common chemical metrics used to evaluate freshwater ecosystem health (Roley et al., 2014). At present, there is a significant deficiency in current understanding of the role of reservoirs as mechanisms of N retention, cycling and removal in peat-dominated catchments where high atmospheric N deposition coincides with a high density of reservoir impoundments supplying major urban areas, for example in northern England. ...
Article
Full-text available
This study presents input-output budgets of total dissolved nitrogen (TDN), dissolved organic N (DON) and dissolved inorganic N (DIN) for a reservoir in a peatland catchment in the south Pennines (UK). This site receives high levels of atmospheric inorganic N deposition, in the range of 26kgNha-1 yr-1. The results show that the reservoir retains ~21 to 31% of the annual TDN input (8806±741kgN). Approximately 39 to 55% of DON (3782±653kgN) and 6 to 13% of DIN (5024±349kgN) were retained/processed. A long water retention time (104days), average annual pH of 6.5, high concentrations of DIN in the reservoir water and a deep water column suggest that denitrification is potentially a key mechanism of N retention/removal. The results also demonstrate that DON is potentially photodegraded and utilized within the reservoir, particularly during the summer season when 58 to 80% of DON input (682±241kgN) was retained, and a net export of DIN (~34kgN) was observed. The findings therefore suggest that DON may play a more crucial role in the biogeochemistry of peat-dominated acid sensitive upland freshwater systems than previously thought. Reservoirs, impoundments and large lakes in peatland catchments may be important sites in mediating downstream N transport and speciation.
Article
Full-text available
1. Wetlands are ecologically and economically important ecosystems but are threatened globally by many forms of human disturbance. Understanding the responses of wetland species to human disturbance is essential for effective wetland management and conservation. 2. We undertook a study to determine (i) whether anurans can be used effectively to assess the ecological integrity of wetlands affected by groundwater withdrawal and, if so, (ii) what effect increasing urbanization might have on the utility of anurans as wetland indicators. We monitored the intensity of anuran calls at 42 wetlands in south-western Florida throughout 2001-2002 and 2005-2009. 3. We first validated the use of anurans to assess wetland integrity using a small group of wetlands by comparing anuran calling and subsequent tadpole development with an established index employing vegetation composition and structure. We then verified that the results could be expanded to a variety of sites throughout the region. Finally, we focused on urbanized wetlands to determine whether urbanization could interfere with the use of anurans to assess wetland integrity. 4. We used PRESENCE to estimate occupancy and detection probabilities and to examine the relationship between occupancy and five covariates expected to influence individual species occurrence. We used FRAGSTATS to calculate the mean proximity index for urbanized wetlands, which assesses the size and distribution of land use types within a specified area. 5. Our results showed that the group of species including oak toad Anaxyrus quercicus, southern cricket frog Acris gryllus, pinewoods treefrog Hyla femoralis, barking treefrog Hyla gratiosa, and little grass frog Pseudacris ocularis is a reliable indicator of wetland integrity. However, this same group of species, which is sensitive to wetland health, is selectively excluded from urbanized wetlands. 6. Synthesis and applications. Although anurans are effective indicators of wetland health and complement vegetation surveys, the usefulness of this group for monitoring the ecological integrity of wetlands can be substantially reduced, or eliminated, as a consequence of urbanization. We urge for careful consideration of confounding factors in any studies examining the utility of indicator species.
Article
Full-text available
Water availability on the continents is important for human health, economic activity, ecosystem function and geophysical processes. Because the saturation vapour pressure of water in air is highly sensitive to temperature, perturbations in the global water cycle are expected to accompany climate warming. Regional patterns of warming-induced changes in surface hydroclimate are complex and less certain than those in temperature, however, with both regional increases and decreases expected in precipitation and runoff. Here we show that an ensemble of 12 climate models exhibits qualitative and statistically significant skill in simulating observed regional patterns of twentieth-century multidecadal changes in streamflow. These models project 10-40% increases in runoff in eastern equatorial Africa, the La Plata basin and high-latitude North America and Eurasia, and 10-30% decreases in runoff in southern Africa, southern Europe, the Middle East and mid-latitude western North America by the year 2050. Such changes in sustainable water availability would have considerable regional-scale consequences for economies as well as ecosystems.
Chapter
The purpose of the Great Lakes Water Quality Agreement (GLWQA) between the U.S. and Canada is to restore and maintain the chemical, physical, and biological integrity of the waters of the Great Lakes basin ecosystem (U.S. and Canada, 1987). The U.S. and Canada have spent billions of dollars and uncounted hours attempting to reverse the effects of cultural eutrophication, toxic chemical pollution, overfishing, habitat destruction, introduced species, and other insults to the ecosystem. Environmental management agencies are now being asked to demonstrate that past remediation programs have been successful and that the results of future or continuing programs will be commensurate with the resources expended (financial and personnel time). The demand for high quality data is forcing environmental and natural resource agencies, which operate with limited resources, to be more selective and more efficient in the collection and analysis of data.
Book
This revised and updated edition of the bestselling Methods in Stream Ecology reflects the latest advances in the technology associated with ecological assessment of streams. In this second edition, all chapters have been updated and modified to reflect the most contemporary protocols covering 6 vital areas of stream ecology: Physical Stream Ecology; Material Transport, Uptake, and Stora Stream B Community Interactions; Ecosystem Processes; and Ecosystem Quality. Each chapter contains basic methods suitable for teaching undergraduate or graduate students and advanced methods for conducting state-of-the-art research. Suitable as a textbook for a course in stream or river ecology, this book is also a critical reference for professional aquatic ecologists, natural resource managers, and for those entering the field of stream ecology who wish to evaluate the condition of streams or their watersheds.
Article
The approximately 11,000 inland lakes in Michigan are valued ecosystems yet are susceptible to degradation due to anthropogenic stresses. Few, if any long-term monitoring programs have been implemented in the inland lakes of Michigan by state agencies. However, Michigan has a lake-volunteer sampling program, the Cooperative Lakes Monitoring Program (CLMP). Our first objective was to assess statewide water quality trends from the early 1970s to present using these volunteer data. For this analysis, we used 71 inland lakes that were distributed across the state that had volunteer-collected Secchi depth (SD). Water clarity in most of these lakes has either increased or stayed the same since the 1970s. Thirty-one percent of the lakes significantly increased in water clarity, 63% had no significant trend and 6% significantly decreased in water clarity. Our second objective was to examine the relationship between lake water clarity and ecoregion and land use/ cover. For this objective, we analyzed 54 lakes from the CLMP program during a time period from 1974-1983 for which we had land use data using t-tests, regressions and analysis of covariance. The mean SD was significantly lower for the southern ecoregion than the northern ecoregion, but we detected few significant relationships between land use/cover and water clarity across lakes. Volunteer monitoring programs provide an invaluable contribution to water quality information and can assist in setting priorities for statewide lake monitoring and management.
Article
Evidence from whole lake experiments is used to reexamine the nutrient control question with particular attention to factors that might be missing in experiments on a smaller scale. Evidence for carbon and nitrogen that have received wide attention as alternatives to phosphorus as limiting nutrients is considered. The control of phytoplankton populations in lakes by carbon and by nitrogen is discussed. The results of studies show that there are biological mechanisms in lakes which are capable of correcting algal deficiencies of carbon, and in some cases, nitrogen. The ratios of carbon to phosphorus and nitrogen to phosphorus are maintained at 174 and 31, respectively. It is suggested that the schemes for controlling nitrogen input to lakes may actually affect water quality adversely by causing low N/P ratios, which favor the vacuolate, nitrogen fixing blue gree algae that are most objectionable from a water quality. When P control causes an increase in the N/P ratio, the resulting shift from water bloom blue green algae to forms that are less objectionable may be as important as quantitative decreases in the algal standing crop.
Article
The suitability of caffeine as a chemical marker for surface water pollution by domestic wastewaters was assessed in this study. Caffeine concentrations in influents and effluents of Swiss wastewater treatment plants (WWTP's, 7-73 and 0.03-9.5 mug/L, respectively) indicated an efficient elimination of 81-99.9%. Corresponding loads in untreated wastewater showed small variations when normalized for the population discharging to the WWTPs (15.8 +/- 3.8 mg person(-1) d(-1)), reflecting a rather constant consumption. WWTP effluent loads were considerably lower (0.06 +/- 0.03 mg person(-1) d(-1)), apart from installations with low sludge age (less than or equal to5 d, loads up to 4.4 mg person(-1) d(-1)). Despite the efficient removal in most WWTPs, caffeine was ubiquitously found in Swiss lakes and rivers(6-250 ng/L, except for remote mountain lakes (<2 ng/L; analytical procedure for wastewater and natural waters: SPE, GCMS-SIM or GC-MS-MS-MRM, internal standard C-13(3)-labeled caffeine). Caffeine concentrations in lakes correlated with the anthropogenic burden by domestic wastewaters, demonstrating the suitability of caffeine as a marker. A mass balance for Greifensee revealed that;: approximate to1-4% of the wastewaters had been discharged without treatment, presumably on rainy days when the capacity of WWTPs had been exceeded. For Zurichsee, it could be shown that the monthly inputs of caffeine correlated with precipitation data. The depth- and seasonal-dependent concentrations in this lake were adequately rationalized by a numerical model considering flushing, biodegradation, and indirect photodegradation via HO. radicals as elimination processes and caffeine inputs as fitting variables.