ESTUARINE INDICATORS - PART 7 (end) potx

Karr et al., 1987 defined biological integrity as “the ability to support and maintain … abalanced, integrated, adaptive community of organisms having a species composition, diversity, a

Trang 1

30

Indicators of Ecosystem Integrity for Estuaries

Stephen J Jordan and Lisa M Smith

CONTENTS

Introduction 467

Terminology and Definitions 468

Conceptual Models 469

Examples of Indicators 471

Water and Sediment Quality Indicators 471

Single-Species (Population-Level) Indicators 472

Community Indicators 472

Ecosystem Indicators 473

Discussion 475

Conclusions 478

Acknowledgments 478

References 478

Introduction

Why do we need indicators of ecosystem integrity for estuaries, what are they, and how would one or more of these indicators inform and contribute to management of these diverse, complex systems?The importance of estuaries was expressed nicely by Welsh (1984, p xiii): “Estuaries … are one of the most heavily utilized and most productive zones in our planet Their integrative processes … weave a web of complexity far out of proportion to their occupation of less than 1% of the planet’s surface.” Citizens groups, environmental managers, and elected officials want to know the status of estuarine ecosystems, locally, regionally, nationally, and globally Especially where there have been large public investments

in pollution controls and other preventive and restorative measures, people want to know if their money has been well spent (Jordan and Vaas, 2000) Thus, there is a clear need for indicators that will provide comprehensive answers to these questions at appropriate intervals Murawski (2000, p 655) recom-mended “simple, robust indices of ecosystem state that gauge … production, diversity, and variability,” and emphasized that indicators should have the capacity to predict the results of management Although Murawski was writing in the context of fisheries and fisheries management, these principles apply more broadly to the integrity of ecosystems in general

The problem, then, is to formulate indicators that are simple in presentation and interpretation They should be robust (i.e., not sensitive to small perturbations or irrelevant factors) and predictive, but also grounded in the complexity and variability that are essential properties of the ecosystem In terms of time and effort, organization of data, and realism in representing the system, such indicators are intermediate between simple descriptive statistics and complex, process-oriented mathematical models

In this chapter we offer several relevant concepts and a few examples of indicators of the large-scale structure and behavior of estuarine ecosystems We also outline principles for development and appli-cation of indicators

2822_book.fm Page 467 Friday, November 12, 2004 3:21 PM

Trang 2

468 Estuarine Indicators

Terminology and Definitions

Indicators can be constructed at various scales of organization, from single chemical, biochemical, orphysiological measurements to highly integrated composites of system attributes (Figure 30.1) Thischapter is concerned principally with the integrative, value-oriented indicators located in the lower rightquadrant of Figure 30.1 Indicators at molecular, sub-organismal, organismal, population, and communitylevels of organization can be informative for various purposes, but indicators at the whole-ecosystemlevel are essential for managing ecosystems and answering the public’s most basic questions

The terms “integrity” and “health” have been used widely to designate desirable states of ecosystems.The goal of the U.S Clean Water Act is to “restore and maintain the chemical, physical, and biologicalintegrity of the Nation’s waters” (U.S Code 33:26:1251a), but the act does not attempt to defineintegrity Without further definition, neither health nor integrity is a useful or measurable descriptor

of an ecosystem A dictionary (Davies, 1976, p 370) defines integrity as “soundness; completeness;unity” and health as “the state of an organism with respect to functioning, disease, and abnormality

at any given time … optimal functioning with freedom from disease and abnormality … flourishingcondition; vitality.” Extending these definitions to an ecological context, integrity might imply thatthe structural properties (state variables) of a system exhibit an expected, undisturbed condition; healthmight be used to indicate how well the system is functioning (rate variables) with respect to expec-tations Karr et al., (1987) defined biological integrity as “the ability to support and maintain … abalanced, integrated, adaptive community of organisms having a species composition, diversity, andfunctional organization comparable to that of natural habitat of the region.” This definition is prob-lematic for estuaries given their complexity, open boundaries, and almost universal lack of “naturalhabitat.” “Condition” is a more general, less value-laden term used in some publications (e.g., U.S.EPA, 2001; Vølstad et al., 2003) in reference to the status of ecosystems Condition, along with thefirst definition of health (“the state of an organism … at any given time”), implies a continuum,whereas the other definitions — along with the definition of integrity — are categorical: one is healthy

or not; one has integrity or does not Each of these words can have inappropriate connotations fornonspecialists Integrity has moral implications Health can be consciously or subconsciously associ-ated with diseases and toxicity Condition may suggest negative status unless modified with positiveadjectives (good condition, excellent condition)

One of the problems with terminology is the attempt to express several properties in a single word.Chesapeake Bay Program (CBP, 1987) policy-makers wrote, “the entire system must be balanced,healthy, and productive.” This statement contains more information than the individual terms discussedabove, and reaches the center of what we are trying to express with ecosystem indicators “Balanced,”

FIGURE 30.1 A hierarchy of indicators.

Trang 3

Indicators of Ecosystem Integrity for Estuaries 469

“healthy,” and “productive” were specifically defined in the context of the Chesapeake restoration in

a later report (Table 30.1) The phrase “the entire system” is also important, suggesting a need forindicators that apply simultaneously and comprehensively to the whole system in all its vastness andcomplexity

Apparently, it is a relatively new idea to capture the essence of large, complex ecosystems in a singleindicator or small set of indicators Only in recent years have monitoring programs begun to generatethe necessary data, and then only for a select, few systems We should expect that less ambiguousterminology will arise, but for now, health, integrity, and condition are used as mostly interchangeabledescriptors of what we are attempting to quantify with these indicators In this chapter, we use the termintegrity, more for consistency with recent literature than by preference

Conceptual Models

Conceptual models are fundamental to indicator development (Boyle et al., 2001) Conceptual models

of ecosystems can have many forms, ranging from minutely detailed flow diagrams based on energy ormaterials, to highly aggregated box models, to qualitative descriptions of expectations or values Typi-cally, conceptual models portray ecosystems as arrangements of interacting parts (molecules, species,trophic guilds, landscape, or seascape mosaics) These concepts lead naturally to indicators based on asuite of interactions, or relationships between inputs and outputs There is an alternative route tostressors through intricate pathways, it is possible to integrate system responses into an indicator ofecosystem integrity Relationships of magnitude and variation between stressors and the indicator, andits component responses can be used to make causal inferences and form testable hypotheses, if thoseare our concerns

A useful type of conceptual model considers the ecosystem as a unitary whole that responds predictablysystem components interact, but how best to represent the system as a whole That is, what indicator orset of indicators accurately tracks changes in the integrity of the ecosystem? How does one quantify theresponse axis of Figure 30.3 and Figure 30.4? The response variable, an indicator of ecosystem integrity,should answer important, comprehensive questions about ecosystems:

What is the status of the ecosystem with respect to one or more reference points?

What is the current direction of change?

How will the ecosystem change in response to external forces, especially management actions?How long will it take to reach a desired or stipulated level of integrity?

TABLE 30.1

Definitions of the Terms Balanced, Healthy, and Productive, as They Apply to Estuarine Ecosystems

Balanced

“Having sufficient populations of prey species to support the species at the top of the food chain, and to limit overabundance

at the bottom of the food chain; no major function of the ecosystem dominates the others”

indicators, as portrayed in the upper oval in Figure 30.2 Rather than tracing the effects of multiple

to stress (Figure 30.3) and changes over time (Figure 30.4) In this conceptual mode, we ask not how

Trang 4

FIGURE 30.2 The contrast between process-oriented models and holistic ecosystem analysis Indicators of ecosystem integrity are derived from integrated system responses as depicted in the upper part of the diagram (The flow diagram at the bottom was adapted, with permission, from an ecosystem model developed by Dan Campbell.)

FIGURE 30.3 Conceptual model of the relationship between ecosystem integrity and stress on the ecosystem On a relative scale, health is the complement of impairment.

Trang 5

Examples of Indicators

The following examples of indicators begin with physical and chemical measurements, and then proceedorganization At each level, we discuss some of the merits and concerns associated with the indicators

Water and Sediment Quality Indicators

Water quality is the most traditional indicator of estuarine condition The ecological and aestheticproblems associated with anoxia, turbidity, eutrophication, and bacterial pollution are obvious even tononscientists (dead fish, brown water, excessive algal growth, human illnesses) Even though theseproblems have been recognized in some estuaries for many decades, the development of water qualitycriteria, standards, and pollutant load capacities specific to estuaries is in its infancy The implications

of using a suite of water quality indicators for the ecological integrity of estuaries is not always clear.The U.S Environmental Protection Agency (U.S EPA) Environmental Monitoring and AssessmentProgram (EMAP) developed indicators of water quality based on dissolved oxygen, chlorophyll a,nutrient concentrations, and water clarity to assess eutrophication Although EMAP is designed to makeecological assessments over large spatial and temporal scales, the data are used to determine conditionfor geographic regions rather than ecosystems Water quality measurements cannot stand alone asindicators of the integrity of an entire ecosystem, because they merely “brush the surface” of biologicalintegrity and ecosystem value We note that EMAP also samples fish and benthic communities, andemploys indicators of biological community structure and health in its assessments

As with water quality, indicators of sediment quality offer only a piece of the puzzle Many sedimentcontaminants are persistent and may be associated with degraded benthic communities, but no criteriafor contaminants in estuarine sediments have been established, only guidelines (Long and Morgan, 1990).Contaminants interact with sediment constituents in ways that can greatly affect their biological avail-ability; thus, sediment concentrations may not be directly correlated with toxicity and biological effects(Bayne et al., 1985) Recent sediment research in estuaries has focused not on contaminants, but onorganic constituents (e.g., pollen and diatom frustules) in benthic cores as paleological records ofecosystem responses to anthropogenic stressors (Bianchi et al., 2000; Dell’Anno et al., 2002)

The National Coastal Condition Report (U.S EPA, 2001) combined indicators of water quality,sediment quality, fish tissue contaminants, wetland loss, and benthic communities into a single index todetermine the overall condition of the nation’s estuaries This method improved multi-indicator integra-tion, but its lack of a strong biological basis limited its interpretation with respect to ecosystem integrity

FIGURE 30.4 Integrity of a stressed ecosystem in the time domain (see also Cairns et al., 1992) (From Jordan, S J and

P A Vaas 2000 Environmental Science and Policy 3:S59–S88 With permission.)

according to the hierarchy depicted in Figure 30.1, ending with an example at the ecosystem-level of

Trang 6

Nevertheless, the application of this index at regional and national scales was an improvement inintegrated assessments

Single-Species (Population-Level) Indicators

Abundance or life history traits (recruitment, growth, mortality) of widely distributed, abundantspecies, or perhaps less abundant, environmentally sensitive species, could be candidates for indicators

of ecosystem health This idea is particularly attractive when the species is well known to the public,for example, popular sport fishes or important commercial species Bortone (2003) discussed thepotential of spotted seatrout (Cynoscion nebulosus) as an indicator species for southeastern estuaries.Because of their popularity and depleted status, striped bass (Morone saxatilis) and Eastern oysters(Crassostrea virginica) became de facto, although scientifically unreliable, indicators of the health ofChesapeake Bay

Striped bass recruitment in the Chesapeake displayed a long-term pattern of increase, followed by a

be explained by changes in fisheries and fishery management, but the pattern was also curiously similarmanager once asserted that the striped bass population was the only indicator needed to track theChesapeake restoration — if the fish recovered, the system would recover Subsequently, the stripedbass population recovered within a decade after stringent fishery management controls were initiated,but the ecosystem remained in many ways far from its desired state For example, the extent of hypoxicand anoxic water did not decline; seagrass coverage, although increasing, was far less than stipulated

by restoration goals; and the long-term decline of oyster populations continued (CBP, 2003)

These few of many possible examples of single-species indicators illustrate two points First, recovery

of a “flagship” species can be encouraging, but nevertheless may occur within a system that is far out

of balance or far from its desired state Second, fisheries and fishery management are integral components

of estuarine ecosystems; they should not be seen as apart from or irrelevant to concerns about waterquality, habitat conditions, and diversity

Community Indicators

Community-level analysis may be indicative of environmental change caused by single or multiplestressors, and predictive of consequences at the ecosystem level According to Attrill and Depledge(1997), community-level investigations are ecologically relevant because changes in communities can

be extrapolated to the health of the ecosystem through changes in food web structure Seagrasses(submerged aquatic vegetation [SAV]) and benthic macrofauna are the most prevalent community-based

FIGURE 30.5 Maryland index of juvenile striped bass relative abundance 1954–1995 (From Maryland DNR, 2003.)

precipitous decline, followed by dramatic recovery (Figure 30.5) This entire dynamic ultimately could

to the conceptual model of ecosystem stimulation, decline, and recovery shown in Figure 30.4 A senior

Trang 7

indicators of estuarine ecosystem health Long-term changes in seagrass coverage in several estuarieshave been strongly associated with nutrient loading and its indirect effects on the availability of light tothe plants This light limitation may result in shifts from SAV-dominated production to proliferation ofphytoplankton and macroalgae (McClelland and Valiela, 1998) Long-term losses and gains in seagrassesrank highly as indicators of eutrophication, but also have some drawbacks as comprehensive indicators

of ecosystem integrity They are vulnerable to physical disturbances (boating, dredging, and commercialfishing operations), diseases, major storms, and other climatic extremes; thus, interpretation of theirdistribution and abundance can be ambiguous

Infaunal macrobenthic communities have been attractive as indicators largely because of their lack

of mobility This trait makes them reliable indicators of exposure, and susceptible to stressors such

as toxic contaminants and severe hypoxia The structure of these communities, however, is sensitive

to factors not directly related to ecosystem integrity such as sediment grain size and organic content.The patchiness of benthic communities over very small spatial scales also can be a drawback inassessing ecosystem integrity over large areas Ranking and categorical reduction of the data (methodsfor generalizing almost any indicator) have been applied to minimize the problem of patchiness atany scale (U.S EPA, 2001)

Indicators based on fish communities or assemblages have received less attention than seagrasses orbenthos The impracticality of complete sampling of estuarine fish communities, along with the migratorybehavior of many species are universal difficulties There have been several attempts to develop estuarineindices of biotic integrity (IBI; Karr et al., 1987) analogous to those used in freshwater systems Examplescan be found in Hughes et al (2002) and Jordan et al (1991) Although IBI approaches are feasible,IBIs for estuarine systems tend to be specific to particular habitats, showing less spatial generality andsensitivity to multiple stressors than might be desired

Vaas and Jordan (1990) used long-term data from seine surveys in the Maryland portion of ChesapeakeBay (Maryland DNR, 2003) to indicate changes in ecosystem integrity They portrayed graphically 3-year means of relative abundance at intervals of decades A simple model based on management goalsings shown in Figure 30.6, developed by Vaas and Jordan (1990) from life history information and clusteranalysis of the seine data, showed an interesting and by now familiar pattern when further synthesizedCommunity indicators are generally more robust than single-species indicators, because they integrateresponses over broader sectors of the ecosystem and a wider range of environmental influences Fishcommunity indicators, for example, would be less sensitive to a single fishery management decisionthan the single-species indicators described above

Ecosystem Indicators

Indicators at the ecosystem level of organization integrate data over biotic communities and trophiclevels, and may include abiotic components (e.g., measures of water quality and physical habitat) Indicessuch as mean trophic level, system-level trophic transfer efficiency, capacity, ascendancy, and overheadhave been used to characterize and compare estuarine ecosystems (e.g., Baird and Ulanowicz, 1989).These types of indices are based on flows of materials (usually carbon) or energy within and throughthe system A stressed ecosystem, for example, might exhibit lower values for mean trophic level, transferefficiency, and ascendancy, along with higher overhead, than an unstressed system In simpler terms, astressed system would be less organized and less efficient, while dissipating energy and materials morerapidly (relative to their supply) than an unstressed system Such indices are valuable tools for gainingunderstanding about the status and relative functions of ecosystems Their principal weakness is thatthey are mathematical abstractions

Multivariate analysis of ecosystem attributes has been used to organize environmental data Thespatial relationships of these attributes can then be used to develop a relative indication of ecosystemintegrity Jordan and Vaas (2000) used cluster analysis of 12 metrics (selected by screening many

candidate metrics) in developing an index of ecosystem integrity for Chesapeake Bay tributaries (Table

and covariation among species predicted future community structure (Figure 30.6) The tolerance

group-for this chapter (Figure 30.7)

Trang 8

ranging from phytoplankton and SAV to fish The index included four water quality metrics in addition

by the Chesapeake Bay Program, so that both societal and ecological values were represented Thefinal index was constructed by ranking six clusters of observations (spanning 9 years and 40 to 50proportions of land cover: urban land predicted low ecosystem integrity, and forested land predictedhigh ecosystem integrity The index was further collapsed into three nominal categories of ecosystemintegrity by calculating the mean cluster value for each site over 9 years, and assigning “good,” “fair,”and “poor” designations to the each third of the distribution of means This procedure produced a

FIGURE 30.6 Observed and predicted changes in relative abundance of 19 species of fish from Maryland Chesapeake Bay seine surveys Data in the top four bar graphs are 3-year averages Note the apparent disruption of the community in the two middle graphs, and the predicted partial recovery in 2000 (Adapted from Vaas and Jordan, 1990.)

30.2) The analysis included a broad array of metrics representing communities and trophic levels

to the biotic variables Six of the metrics were normalized to restoration goals previously established

sites) into an ordinal scale of ecosystem integrity (Figure 30.8) The index was sensitive to watershed

simple display of long-term, large-scale geographic patterns (Figure 30.9)

Trang 9

Discussion

Ecosystem integrity is a human construct that defies rigorous scientific definition To assess the integrity

of an ecosystem requires reference points defined by humans, who lack perfect knowledge of the system’sstructure and functions Because we cannot determine from first principles “what is a good or badecosystem,” we ask reasonable people what they want from the system For Chesapeake Bay, the desired

FIGURE 30.7 Mean rank abundance of tolerant species in Maryland Chesapeake Bay seine surveys, 1960–2000 Lower values indicate higher abundance of tolerant species and lower ecosystem integrity.

TABLE 30.2

Metrics Used to Construct an Index of Ecosystem Integrity for Chesapeake Bay

Submersed Aquatic Vegetation Success

Percentage of potential habitat vegetated (+) Batiuk et al., 1992

Deviations from Water Quality Goals

Dissolved inorganic nitrogen (NO2 + NO3 + NH4) (–) Batiuk et al., 1992 Dissolved inorganic phosphorus (–) Batiuk et al., 1992 Secchi depth (+) Batiuk et al., 1992 Chlorophyll a (–) Batiuk et al., 1992

Plankton

Biomass of nuisance algal species (cyanophytes and dinoflagellates) (–) Jordan and Vaas, 2000 Ratio of mesozooplankton to microzooplankton abundance (+) Buchanan et al., 1993 Biomass of microzooplankton (–) Buchanan et al., 1993

Note: Plus (+) and minus (–) signs indicate positive or negative relationship to ecosystem integrity.

Source: Adapted from Jordan and Vaas (2000).

Trang 10

states are health, balance, and productivity (CBP, 1987), as defined in the introduction to this chapter

A more recent document (CBP, 2000) explicitly acknowledged the human and conceptual elements incalling for “a shared vision — a system with abundant, diverse populations of living resources,” andreiterated the need for the entire system to be healthy and productive

The examples of indicators presented here all reflect one or more elements of balance, health, orproductivity The striped bass index is a strong indicator of productivity, but tells little about health orbalance In fact, as the Chesapeake striped bass population recovered and became very abundant in thelate 1990s and early 2000s, it became less healthy, displaying consistent, abnormally high prevalence

of starvation and disease (Jacobs et al., 2002) Perhaps in addition to the obvious health implications,these problems were symptomatic of an unbalanced condition, with overproduction of an importantpredator

The Chesapeake Bay fish community indicators include elements of balance (species dominance),productivity (abundance), and, less directly, health (tolerance groups) They lack reference to values orreasoned expectations, however Instead, they depend on an internal historical reference (the structure

of the community ca 1960) The pristine condition could not be used as a reference because it was

FIGURE 30.8

higher ecosystem integrity The year 1991 is shown as an example; the analysis spanned 1986–1994 (From Jordan, S J and P A Vaas 2000 Environmental Science and Policy 3:S59–S88 With permission.)

An index of ecosystem integrity based on cluster analysis of 12 metrics ( Table 30.2 ) Higher values represent

Trang 11

unknown Where sufficient data exist, historical references are better than none; they do indeed representvalues when society seeks to restore overly stressed ecosystems to former, less-stressed conditions.The index of ecosystem integrity for Chesapeake Bay (Jordan and Vaas, 2000) includes indicators ofbalance, health, productivity, and diversity Moreover, it is based on values, as 6 of 12 component metricsare normalized to specific, numerical restoration goals for water quality and biotic communities Itsweaknesses include problems with missing data and fixed-station sampling designs that lead to difficulties

in statistical and geographical inferences The potential of applying the index in a predictive mode hasnot been explored fully, although the strong relationship with land cover could be a starting point forpredictive analysis

The scarcity of data for most estuaries is a major impediment to the development and application ofecosystem-level indicators, which require extensive, comprehensive long-term monitoring programs Theefforts of the National Coastal Assessment (U.S EPA, 2001) to establish consistent monitoring anduniversal indicators for U.S estuaries are improving the quantity, quality, and availability of data relevant

to ecosystem integrity The program would provide more support for indicators of ecosystem integrity

by monitoring a broader spectrum of biological responses (plankton and aquatic vegetation in addition

to fish and benthic communities) Longevity will be crucial In large estuarine systems, ecosystem-levelresponses to changes such as contaminant load reductions or land management practices can take years

or decades to become detectable Therefore, monitoring programs must be sustained indefinitely Thefish community indicators described here were possible only because the Maryland seine surveysgenerated a consistent record extending over several decades The longevity of a monitoring program is

FIGURE 30.9

for each site over 9 years (1986–1994) were collapsed into three categories of ecosystem integrity (From Jordan, S J and

P A Vaas 2000 Environmental Science and Policy 3:S59–S88 With permission.)

Long-term ecosystem integrity of northern Chesapeake Bay Mean cluster values (1–6; see Figure 30.8 )

Trang 12

We thank the many people and organizations that contributed their work, data, thoughts, and resources

to the ideas and examples presented in this chapter Naming them all would require yet another chapter.The information in this document has been funded in part by the U.S Environmental Protection Agency

It has been subjected to review by the National Health and Environmental Effects Research Laboratoryand approved for publication Approval does not signify that the contents reflect the views of the agency,nor does mention of trade names of commercial products constitute endorsement or recommendationfor use This is Contribution 1194 of the Gulf Ecology Division

Batiuk, R P., R Heasley, R Orth, K Moore, J Capelli, J C Stevenson, W Dennison, L Staver, V Carter,

N Rybicki, R E Hickman, S Kollar, and S Beiber 1992 Chesapeake Bay Submerged AquaticVegetation Habitat and Restoration Goals: A Technical Synthesis U.S Environmental Protection Agen-

cy Chesapeake Bay Program Report, Annapolis, MD

Bayne, B L., D A Brown, K Burns, D R Nixon, A Ivanovici, D R Livingstone, D M Lowe, M N.Moore, A R D Stebbing, and J Widdows 1985 The Effects of Pollution on Marine Animals PraegerScientific, New York, 384 pp

Bianchi, T., P Westman, C Rolff, E Engelhaupt, T Andrén, and R Elmgren 2000 Cyanobacterial blooms

in the Baltic Sea: natural or human-induced? Limnology and Oceanography 45:716–726

Bortone, S A (ed) 2003 Biology of the Spotted Seatrout. CRC Press, Boca Raton, FL

Boyle, M., J J Kay, and B Pond 2001 Monitoring in support of policy: an adaptive ecosystem approach

In Encyclopedia of Global Environmental Change, Vol 4, T Munn (ed.) John Wiley & Son, New York,

pp 116–137

strongly related to the potential for robust indicators and predictive models Taken together, Figure 30.5

through Figure 30.7 illustrate the relationships between short-term interannual variability, which could

Trang 13

Buchanan, C., R W Alden III, R S Birdsong, F Jacobs, and K G Sellner 1993 Development of ZooplanktonCommunity Environmental Indicators for Chesapeake Bay: A Report on the Project’s Results throughJune 1993 Interstate Commission on the Potomac River Basin, Rockville, MD

Cairns, J., J R McCormick, V Paul, and B R Niederlehner 1992 A proposed framework for developingindicators of ecosystem health Hydrobiologia 263:1–44

Carmichael, J B., B Richardson, M Roberts, and S J Jordan 1992 Fish Sampling in Eight ChesapeakeBay Tributaries Technical Report, Chesapeake Bay Research and Monitoring Division, MarylandDepartment of Natural Resources, Annapolis

CBP 1987 Chesapeake Bay Agreement Chesapeake Bay Program, Annapolis, MD Available atCBP 1993 Chesapeake Bay Strategy for the Restoration and Protection of Ecologically Valuable Species.Chesapeake Bay Program CBP/TRS 113/94, Annapolis, MD

CBP 2000 Chesapeake 2000: A Watershed Partnership Chesapeake Bay Program, Annapolis, MD Available

Davies, P (ed.) 1976 The American Heritage Dictionary of the English Language. Dell, New York.Dell’Anno, A., M L Mei, A Pusceddu, and R Danovaro 2002 Assessing the trophic state and eutrophication

of coastal marine systems: a new approach based on the biochemical composition of sediment organic

Jordan, S., P Vaas, and J Uphoff 1991 Fish assemblages as indicators of environmental quality in northernChesapeake Bay In Biological Criteria: Research and Regulation 1990 Proceedings of a Symposium.U.S Environmental Protection Agency Office of Water, Washington, D.C

Jordan, S J., C Stenger, M Olson, K Mountford, and R Batiuk 1992 Dissolved Oxygen Restoration Goalsfor the Chesapeake Bay Living Resources U.S Environmental Protection Agency Chesapeake BayProgram, Annapolis, MD

Karr, J R., P R Yant, and K D Fausch 1987 Spatial and temporal variability of the index of biotic integrity

in three midwestern streams Transactions of the American Fisheries Society 116:1–11

Long, E R and L G Morgan 1990 The Potential for Biological Effects of Sediment-Sorbed ContaminantsTested in the National Status and Trends Program NOAA Technical Memorandum NOS OMA 52,Rockville, MD

M a r y l a n d D N R 2 0 0 3 S t r i p e d B a s s S e i n e S u r vey J u ve n i l e I n d ex Pa g e Ava i l a b l e a tMcClelland, J W and I Valiela 1998 Changes in food web structure under the influence of increasedanthropogenic nitrogen inputs to estuaries Marine Ecology Progress Series 168:259–271

Murawski, S.A 2000 Definitions of overfishing from an ecosystem perspective ICES Journal of Marine Science 57:649–658

Ranasinghe, J A., S B Weisburg, J B Frithsen, L C Schaffner, R J Diaz, and D M Dauer 1993.Chesapeake Bay Benthic Community Restoration Goals Technical Report, prepared for The ChesapeakeBay Program Office and The Governor’s Council on the Chesapeake Bay Versar, Inc., Annapolis, MD.Rapport, D J 1999 On the transformation from healthy to degraded aquatic ecosystems Aquatic Ecosystem Health and Management 2:97–103

U.S EPA 2001 National Coastal Condition Report U.S EPA Office of Research and Development, Vaas, P A and S J Jordan 1990 Long term trends in abundance indices for 19 species of Chesapeake Bayfishes: reflection of trends in the bay ecosystem In New Perspectives in the Chesapeake System:

EPA-A Research and Management Partnership Proceedings of a Conference Chesapeake Research tium Publication 137

Consor-2822_book.fm Page 479 Friday, November 12, 2004 3:21 PM

http://www.chesapeakebay.net/pubs/1987ChesapeakeBayAgreement.pdf

at http://www.chesapeakebay.net/agreement.htm

CBP 2003 Bay Trends and Indicators Chesapeake Bay Program Web page:

http://www.chesapeakebay.net/in-matter Marine Pollution Bulletin 44:611–622.

Trang 14

Vølstad, J H., N K Neerchal, N E Roth, and M T Southerland 2003 Combining biological indicators ofwatershed condition from multiple sampling programs — a case study from Maryland, USA Ecological Indicators 3:13–025

Welsh, B L 1984 Foreword In The Estuary as a Filter, V S Kennedy (ed.) Academic Press, New York,

511 pp

Trang 15

31

Using the Human Disturbance Gradient to Develop Bioassessment Procedures in Estuaries

Ellen McCarron and Russel Frydenborg

CONTENTS

Introduction 481

Methods 482

The Human Disturbance Gradient 483

Selection of Biological Metrics for the Aggregate Stream Condition Index 484

Results 484

Discussion: Developing Bioassessment Criteria for Estuaries 484

Classifying Estuaries — Developing Reasonable Expectations and Making Valid Comparisons 485

Determining the Potential Elements of an Estuarine Human Disturbance Gradient 487

Sampling Biota across a Gradient of Human Disturbance 488

Examining Biological Attributes Associated with Estuarine Community Health 489

Selecting Community Attributes Associated with Human Disturbance 490

Communicating the Results Clearly to Stakeholders and Policy Makers 490

Conclusions 490

Acknowledgments 491

References 491

Introduction

Storm water and other non-point sources of pollution are major contributors to the degradation of surface water and groundwater resources (FDEP, 1999) In recognition of this disturbing trend, in the early 1980s the Florida Department of Environmental Protection (FDEP) developed and implemented what is now

a nationally recognized program to manage non-point source pollution One of the main difficulties facing this newly formed program was the lack of appropriate techniques to monitor and assess surface water impairment The traditional chemistry-based analysis of water samples was inadequate because

of the inherently transient and unpredictable nature of non-point pollution

An alternative approach, developed at the Department in the late 1980s and early 1990s, provided an ecologically based solution to the problem of monitoring non-point sources The solution was the development and implementation of community-level biological assessment (bioassessment) tools using multiple metrics (a multimetric index) Each of these metrics is an attribute that responds in a predictable manner to human disturbance

The conceptual approach for developing a multimetric index was proposed by Karr and Chu (1999) Completing this process for streams in Florida has taken well over a decade Resource limitations, as well as the nature of emerging national guidance from the U.S Environmental Protection Agency (U.S

Trang 16

Florida’s Total Maximum Daily Load Program is one of several high-profile programs in the ment that uses the stream and lake bioassessment tools Recently adopted rules for this program (Rule62-303, Florida Administrative Code) set forth objective criteria for determining impairment and areused to determine which waters will be placed on the state’s official list of impaired waters that issubmitted to the federal government for approval The rules incorporate the Stream Condition Index,BioRecon, and Lake Condition Index as measures of surface water quality impairment By includingthese bioassessment measures in Rule 62-303, the FDEP has greatly enhanced its ability to detectimpairment and restore waters in need of remediation, through its Total Maximum Daily Load Program.The development of a bioassessment tool for estuaries has long been one of the FDEP’s goals Because

Depart-of the inherent complexities Depart-of estuarine systems and the lack Depart-of any proven national models, however,

an estuarine tool that is applicable statewide has not been developed The recent use of a HumanDisturbance Gradient approach to recalibrate the Stream Condition Index could potentially bring renewedenergy to the development of estuarine bioassessment procedures

The following section on methods describes how the stream bioassessment tool was recalibrated usingthe Human Disturbance Gradient approach The discussion and conclusions sections explore how theHuman Disturbance Gradient theory underlying this important stream recalibration project could provide

a potential framework for developing a bioassessment approach in estuarine systems

Methods

Following national guidance from the U.S EPA (Plafkin et al., 1989) in the early 1990s two streammacroinvertebrate multimetric bioassessment tools were developed in Florida These tools, the StreamCondition Index and BioRecon, were developed, tested, and implemented by comparing the biologicalcondition at sites with minimal human influence (reference sites) and sites with known disturbance (testsites) (Barbour et al., 1996)

In accordance with guidance from the U.S EPA, a geographic framework was also developed for theStream Condition Index and BioRecon tools, because regional differences translate into different expec-tations for the biological communities present For example, there is naturally much higher taxonomicdiversity in northern Florida, compared with central and southern Florida A first attempt at developing

a regional geographic framework used the nationally designated ecoregions in Florida (U.S EPA, 1989)and partitioned them further into subecoregions, using methods similar to those used in mapping thenational ecoregions This project produced a multiuse map of Florida’s subecoregions (U.S EPA, 1994)circularity in the methodology A statistical evaluation of macroinvertebrate data against the subecore-gional framework led to the identification of the three “bioregions” used for the Stream Condition Index

The Stream Condition Index and BioRecon have proved useful in the FDEP’s assessment of Florida’sbiological communities However, the Human Disturbance Gradient approach, which emerged in theearly 1990s, offers a greater level of discrimination between biological condition (y axis) and environ-mental condition (x axis) As a result, the FDEP has just completed contract work with Leska Fore ofStatistical Design, Inc to recalibrate the Stream Condition Index and BioRecon tools using a HumanDisturbance Gradient approach Fore’s report (November 2003) discusses in detail the methods used inthe Stream Condition Index recalibration The following sections summarize the Human DisturbanceGradient theory and how it was applied in recalibrating the stream bioassessment tool

and BioRecon assessment tools, i.e., the Florida panhandle, peninsula, and northeast bioregions (Figure

(Figure 31.1) Biological data were intentionally not used in the development of subecoregions to avoid

31.2)

Trang 17

Using the Human Disturbance Gradient to Develop Bioassessment Procedures in Estuaries 483

The Human Disturbance Gradient

The Human Disturbance Gradient was first introduced by Karr (1993) This landmark publication built

on what Karr and his colleagues described as the five factors that summarize how humans alter anddegrade water resources (Karr et al., 1986, 2000): hydrology, physical habitat structure, water quality,energy source, and biological interactions Data were available in Florida to measure the first four ofthese five factors, as follows

FIGURE 31.1 Subecoregions of Florida.

FIGURE 31.2 Bioregions of Florida.

65f – Southern Pine Plains and Hills 65g – Dougherty/Marianna Plains 65h – Titon Upland/Tallahassee Hills 75a – Gulf Coast Flatwoods 75b – Southwestern Florida Flatwoods 75c – Central Florida Ridges and Uplands 75d – Eastern Florida Flatwoods 75e – Okefenokee Swampls and Plains 75f – Sea Island Flatwoods 76a – Everglades 76b – Big Cypress 76c – Miami Ridge/Atomic Coastal Strip 76d – Southern Coast and Islands

65h 75a

75e 75f

75c

75b

75d

76c 76b

Trang 18

Hydrology. A hydrologic scoring system was developed based on FDEP biologists’ knowledge ofwater removal (drainage, consumptive use), patterns of drought, and hydrographs for the specific sitesused in the recalibration exercise

Physical habitat structure. Habitat assessment criteria were developed for the FDEP’s stream assessments, and a standardized form is always completed for each Stream Condition Index and BioReconassessment The habitat assessment score includes substrate condition and availability, water velocity,habitat smothering, channelization, bank stability, and the width and vegetative condition of the riparianzone

bio-Water quality.Ammonia concentration was used because it had the most complete record of data, has

a high correlation with other water quality measures, and is indicative of degradation from a wider range

of land uses than other measures

Energy source.As a direct measure of energy types and quantities was not available, a surrogate wasselected that is an index of nonrenewable energy flow in the surrounding catchment, or geographic area.This surrogate, the Land Development Intensity Index, was developed by Dr Mark Brown and colleagues

at the University of Florida (Brown and Vivas, in press) It is calculated as the percentage area withinthe catchment of particular types of land uses, multiplied by a coefficient of energy associated with thatland use, and then summed over all land use types

Selection of Biological Metrics for the Aggregate Stream Condition Index

To determine the variety of attributes that is reliably associated with human disturbance, 36 candidatebiological metrics were evaluated using the following criteria: (1) meaningful measure of ecologicalstructure or function; (2) strong correlation with the Human Disturbance Gradient; (3) statistically robust,with low measurement error; (4) reflecting multiple categories of biological organization; (5) cost-effective to measure; and (6) provide information that is not redundant with other metrics

The final metrics selected had all of these characteristics They were then normalized into unitlessscores, using the 95th or 5th percentile values for each metric Each metric value was divided by itsrange and multiplied by 10 For metrics that decreased with an increasing Human Disturbance Gradient,the values were scored 0 to 10, and for metrics that increased with a Human Disturbance Gradient, thevalues were scored 10 to 0, so that in either case, the best metric score was a 10

As described in Fore (2003), this process was carried out separately for each of the three Floridabioregions, and calibration factors were assigned to the metrics as needed to ensure similar responsesfor all metrics in all regions The unitless metric scores were then summed into a single Stream ConditionIndex value for each site Finally, statistical tests were applied to the data set to determine the number

of categories that the Stream Condition Index aggregate index could detect

Results

Ten metrics were selected for inclusion in the aggregate Stream Condition Index, and most of themfound to be a reliable predictor of biological condition, as it was highly correlated with the StreamUsing the statistical tests described in Fore (2003), there are approximately five assessment categoriesThe FDEP’s current rules (Rule 62-303, Florida Administrative Code) require two visits to a site tomake a definitive assessment using the Stream Condition Index

Discussion: Developing Bioassessment Criteria for Estuaries

Using the procedure followed for streams as a template, the following steps can be taken to developeffective bioassessment criteria for estuaries: (1) based on knowledge of a region’s estuaries, properly

required regional scoring calibration (Table 31.1) Also, the FDEP’s Human Disturbance Gradient wasCondition Index (Figure 31.3)

for two visits at the same site and four categories when only one visit at a site is obtained (Table 31.2)

Trang 19

Using the Human Disturbance Gradient to Develop Bioassessment Procedures in Estuaries 485

classify these estuaries into meaningful units; (2) develop criteria for a Human Disturbance Gradientfor estuaries; (3) establish the reference condition, or the biological expectations within each classifiedunit, based on the Human Disturbance Gradient; (4) sample the biota at sites representing conditionsacross the Human Disturbance Gradient, using the same methods and taxonomic resolution; (5) examinebiological attributes associated with estuarine community structure, composition, and life history; (6)determine a variety of attributes that is reliably associated with human disturbance (see the earlierdiscussion on the selection of biological metrics for the Stream Condition Index); and (7) use thisinformation to assess the quality of an estuarine site in a manner that is easily communicated tostakeholders and policy makers (e.g., the multimetric index approach)

Classifying Estuaries — Developing Reasonable Expectations and Making Valid Comparisons

The appropriate classification of estuarine sites is extremely important for establishing biological munity expectations Following the Karr and Chu (1999) model, geographic setting should be the firstelement in the classification scheme Florida contains three marine provinces: the Louisianan, WestIndian, and Carolinian Climate, prevailing currents, the availability of specific habitats, and organismrecruitment broadly influence the composition of biological communities in these provinces

com-Habitat typeis the next logical element in developing a classification system for bioassessment Marinehabitats are generally separated into floral-based, faunal-based, and mineral-based communities

TABLE 31.1

List of Ten Metrics Showing Regional Thresholds

Total taxa 16–42 16–49 16–41 Ephemeroptera taxa 0–3.5 0–6 0–5 Trichoptera taxa 0–6.5 0–7 0–7

% Filterer 1–42 1–45 1–40 Long-lived taxa 0–3 0–5 0–4 Clinger taxa 0–9 0–15.5 0–8

% Dominance 54–10 43–10 54–10

% Tanytarsini 0–26 0–26 0–26 Sensitive taxa 0–11 0–19 0–9

Non-Outlier Range Outliers

100 80 60 40 20 0

Human disturbance gradient

2.8 2.4 2.0 1.6 1.2

(Figure 31.4) Obviously, the biota inhabiting a bed of submerged aquatic vegetation (a floral-based

Tiêu đề	Indicators of Ecosystem Integrity for Estuaries
Tác giả	Stephen J. Jordan, Lisa M. Smith
Trường học	Not Available
Chuyên ngành	Ecosystem Integrity
Thể loại	Chapter
Năm xuất bản	2004
Thành phố	Not Available

Định dạng
Số trang	39
Dung lượng	2,28 MB