PB91-216200
Choices  in Monitoring Wetlands
ManTech Environmental Technology, Inc., Corvallis, OR




Prepared for:

Corvallis Environmental Research Lab., OR




1991
                U.S. DEPARTMENT OF COMMERCE
              National Technical Information Service

-------

-------
                                                         ERL-COR- /3V-;f' £>
                                TECHNICAL REPORT DATA
                                \Imarucne*a on the nvtne be fort eomplttf-
(Hcett mdlmitrucriofu on
 1. REPORT NO.
  EPA/600/D-91/129
  2.
                                                           PB91-216200
 4. TITLE AND SUSTITLE

    Choices in Monitoring Wetlands
                            ft. REPORT DATE
                            k. PERFORMING ORGANIZATION CODE
 7. AUTMOR1S)

    Paul Adamus
                            I. PERFORMING ORGANIZATION REPORT NO.
 •. PERFORMING ORGANIZATION NAME AND ADDRESS
                                                      10. PROGRAM ELEMENT NO.
    Mantech Environmental Technology,  Inc.,
    ERL-Corvallis, OR
                            11. CONTRACT/CHANT NO.
 12. SPONSORING AGENCY NAME AND ADDRESS
  US Environmental Protection Agency
  Environmental  Research Laboratory
  200 SW 35th Street
  Corvallis, OR  97333	
                            13.TVPEOF REPORT AND PERIOD COVERED
                             	Symposium Paper
                            14. SPONSORING AGENCY CODE
                              EPA/600/dz
   SUPJfLJLMENT-ARY NOTES   ,  _  ,.
    19F1.  Ecological  Indicators:  Proceedings of  the USEPA  International
    Symposium,  D. McKenzie and  E.  Hyatt,  eds.
     5TRACT
    Efforts  to  develop  and  compare  indicators  on  wetland  ecological
    condition  should employ designs that span a gradient of disturbed and
    undisturbed   (but  otherwise  as  similar  as possible)  wetlands.   As
    resources  allow, they should compare all taxa and ecosystem processes,
    as well  as  metrics and  data  reduction  techniques,  which  from   a
    theoretical perspective and studies  to  date show promise for use.  they
    should be  regionally-based,  covering specific wetland types as defined
    by predominant hydrologic regime, chemical regime, and vegetation form.
    Empirical  results  should be  integrated with  results from  experiments
    and  simulation models to identify wetland components most  suitable as
    indicators
17.
   KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Croup
                          REPRODUCED SY
                          U.S. DEPARTMENT OF COMMERCE
                          NATIONAL TECHNICAL INFORMATION SERVICE
                          SPRINGFIELD, VA. 22161
1B. DISTRIBUTION STATEMENT
   Release  to Public
                                          IB SICJJRITV. CI.ASS
                                           Unclassified
                                        21. NO. OF PAGES
                                               25
                M. SECURITY CLASS
                  Unclassified
22. PRICE
•PA r»rm uao-t

-------
                                                                        PB91-216200

                                                                      EPA/600/D-91/129
                                                 1
                             CHOICES IN MONITORING WETLANDS

                                        PAUL R. ADAMUS
                              ManTech Environmental Technology Inc.
                                US EPA Environmental Research Lab
                                200 SW 35th St., Corvallis, OR 97333
                                                1.1
                                         INTRODUCTION

Wetlands pose unusual challenges for monitoring programs. The enormous spatial and temporal variability
that is typical of wetlands requires that large numbers of samples be collected if the wetland community is
to be properly characterized.  However, access problems severely limit the ability to easily sample wetlands.

Nonetheless, the need for more vigorous wetland sampling efforts is compelling.  Many undisturbed wetlands
are characterized by  exceptional biological productivity,  but  at  the same  time, can easily accumulate
contaminants from a wide area.  This is partly because most wetlands are located in a topographically low,
depositional environment and have long hydraulic detention times.  This combination of normally great
productivity and large potential for exposure to unnatural stresses suggests a need for extensive monitoring,
so remedial action  may be taken if wetlands  begin  to  show signs of  functional impairment.  However,
wetlands seldom are monitored in a geographically extensive, comparative manner. This chapter begins with
discussions of considerations for wetland monitoring programs, and later discusses differences that may occur
among wetland types and regions, using as an example an  analysis of existing bird databases.


                                                1.2
                                 DECIDING WHAT TO MONITOR

The choice of what to monitor is governed by both policy and scientific considerations.  Monitoring typically
focuses on "indicators" of ecological condition; these may  be physical,  chemical, or biological samples or
measurements of processes. Criteria for evaluating potential indicators - such as sensitivity,  repeatability,
and cost-- have been proposed many times (e.g., Hellawell 1984, Kelly and Harwell 1989, Landres et al. 1988,
Schaeffer et al. 1988). Final selection of indicators requires that the relative weights assigned to various
criteria reflect the potential purposes for which interpretations of the indicator are being used.  For example,
wetland biomonitoring may be conducted to meet any of the following goals:

o       Determining whether a wetland is changing, and in what direction;

o       Assessing how aberrant is the community structure of a particular wetland, e.g., to set priorities for
        restoration or strategies  for mitigation;

o       Evaluating the success of management of a wetland, e.g., compliance with permits and mitigation
        plans;

o       Pinpointing the source of degradation of a wetland;

o       Evaluating overall  program success of wetland quality protection efforts;

o       Priority ranking of wetlands;

-------
o       Gaining an understanding of fundamental wetland processes and advancing science.

As part of its planning for the nationwide Environmental Monitoring and Assessment Program (EMAP),
The U.S. Environmental Protection Agency (USEPA), the U.S. Environmental Protection Agency's (USEPA)
Wetlands Research Program has drafted a list of parameters it believes show the most promise for indicating
the ecological condition of wetland classes at the regional and national level, given an assumption that each
of the approximately 3000 statistically-selected wetlands would be visited only once or twice a year, once
every four years (Table 1). Details of the sampling program, which will be subject to ongoing peer review,
testing, and revision, are contained in Leibowitz et al. (1991).

Ideally, monitoring of a wetland should encompass as long a time period, as many indicators, and as many
microhabitats within the wetland as possible, given available resources.  By monitoring both short- and long-
lived taxa, for example, the  effects both of stressors  that occur briefly (e.g., herbicides) and of those that
occur over longer time periods (e.g., bioaccumulation of metals) can be detected.  By monitoring both
resident and wide-ranging/ migrant species, for example, the cumulative landscape-level impacts that may not
be detectable on a local scale may become apparent.  And by measuring the community-level responses as
well as the individual and population-level responses to a stressor, causal mechanisms become more evident.

If a sampling program cannot be taxonomically comprehensive, three approaches can be used to identify the
taxa  that are most important to sample:
        o       Physiological sensitivity - empirically based
        o       Physiological sensitivity - experimentally based
        o       Functional importance

Selecting Indicators: Empirical Approaches

An empirical approach to indicator selection would focus  on identifying taxa  or communities that are
physiologically sensitive, based on empirical results of field monitoring surveys spanning a gradient of stressor
conditions.  The USEPA has just completed such  a  review  of literature describing such studies in inland
wetlands (Adamus and Brandt 1990).   The report is organized by  major phyla (e.g., microbes through
mammals) and documents the impact on each phyla  of 12 major stressors (Table 2).  Appendices include
a bibliography of wetland biological community for each state, referenced to a map snowing study locations.
Figure 1 shows, on a national scale, the locations of community-level studies of inland  wetlands referenced
in the report.

Although the Adamus and Brandt (1990) review highlighted several potentially suitable indicator taxa, the
studies from which conclusions  might be drawn are not necessarily representative of the wetland population
generally. Different  combinations of stressor magnitude, wetland type, biological  community structure, and
hydrologic regime, for  example, may prohibit generalization  of the conclusions drawn from cited studies.

A few studies cited in Adamus  and Brandt (1990) explicitly compared the relative sensitivities of the major
taxonomic groups, with regard to anthropogenic stress in inland wetlands. For example, Brooks et al. (1990)
compared the relative sensitivities of amphibians, birds, mammals, and aquatic life to alteration of riparian
wetlands in Pennsylvania at a catchment scale. They  concluded that bird community structure was generally
a better indicator of landscape disturbance than was mammal or amphibian community structure. Similarly,
Ohmart et al. (this volume) compared the relative sensitivities of amphibians, birds, mammals, and vegetation
in Arizona riparian systems  to  increased salinity and flow regime alteration. They concluded that surveys
of vegetation composition were a more cost-effective indicator of those types of stressors than was censusing
of vertebrates.

Studies that compare both structural and functional indicators in wetlands are rare. One such study is that
of Aust et al. (1988), examining the effects of silvicultural practices on a North Carolina forested wetland.

-------
They concluded that the "most efficient" indices of changes in ecological function were soil acidity, redox
potential, oxygen concentration, temperature, soil mechanical resistance, sedimentation, and vegetation cover.
These indices require short sampling periods, a minimum of laboratory work,  and easily operable and
maintainable equipment.  "Less complex to  interpret" were sedimentation,  net primary productivity, plant
nitrogen and phosphorus uptake, cellulose decomposition, and bird richness, diversity, and abundance. "Most
responsive to disturbance" (i.e., showing significant differences across gradients or between treatments) were
total nitrogen and phosphorus concentrations in soil water, soil acidity, redox potential, saturated hydraulic
conductivity, tempeiature, soil mechanical resistance, sedimentation, net primary productivity, plant N and
P uptake, and cellulose decomposition.  "Most integrative of ecological processes" were soil redox potential,
net primary productivity, plant N and P uptake, and cellulose decomposition rates.

Selecting Indicators: Experimental  Approaches

A second approach for estimating relative physiological sensitivities  of candidate indicators is to review
literature on experiments in  which wetland communities  were intentionally exposed  to a stressor in an
experimental manner, and subsequently monitored. Few such experiments have been conducted in natural
outdoor settings; most have  involved limited-time, single-species assays of under laboratory  conditions.
Literature describing bioassay results for a very limited number of wetland taxa (almost entirely animals) can
be accessed using several computerized databases, e.g., the USEPA's AQUIRE database (Pilli et al. 1989).

The standardized conditions used  in most toxicity testing allow some degree of comparison  among taxa
regarding their relative sensitivities. However, wetland communities are often stressed to greater degree by
hydrologic perturbations and burial  by sediment,  which have not been the focus  of bioassays, than by
chemical  contaminants.  Even when chemical bioassay data  are available for a wetland organism, the
interpretation can be confounded by differences between laboratory test conditions and field conditions that
are  typical  of wetlands (e.g., altered  toxicant  mobility and  toxicity due  to  increased organic  carbon;
interactions between hydroperiod effects and chemical toxicity). To address  gaps in extant data, additional
field bioassays might be undertaken.  For example, in the case of vascular plants, relative sensitivities might
be elucidated by measuring exposure of a host of species  to a particular substance (e.g., a nutrient) and then
monitoring the varying degrees to which the substance accumulates in tissue (e.g., Canfield  et al. 1983), or
alters germination and other physiological processes.

Selecting Indicators: Functional Importance Approaches

A third approach  for focusing monitoring efforts involves identifying taxa whose ecological roles are
disproportionately great in  comparison to their density.  Some  biologists  have  termed these "keystone"
species.  Particular  species assemblages (functional  groups, or "guilds") may also share similar responses to
a particular stressor, and thus collectively serve as a useful indicator.  These often include taxa,  such as the
following, which physically alter the landscape so profoundly that they create or destroy habitat  for a much
larger group of species over a wide area:

o       Muskrats, alligators, and some herbivorous  birds, which cause locally major increases in  open water
        patchiness of wetlands by their grazing  activities and physical  movement;

o       Beaver, which create wetlands and temporarily destroy forest;

o       Sphagnum mosses, which provide substrate for other plants and animals, and alter wetland hydrology,

o       Common carp (Cvprinus carpio). which extensively resuspends sediments  as  it  feeds;

o       Woodpeckers, which excavate cavities required by dozens  of species;

-------
o       Bees and other pollinating or seed-dispersing organisms, which control  habitat structure through
        their major collective effects on vegetation;

o       Gopher tortoises and other burrowing species that create shelter critical  to survival of many other
        animals.

Considerable prior understanding of ecosystem processes and life histories of prevalent species is needed in
order to utilize a keystone approach to monitoring.   Modeling techniques can be used initially to help
identify sensitive taxa or processes.  Wetland models (e.g., Patterson and Whillans  1984)  describe the
connections among ecosystem compartments, e.g., species. When many connections converge on a particular
compartment,  or  when (through  mathematical  simulations)  it becomes  apparent  that  a particular
compartment exerts considerable influence on other compartments, it may be tentatively assumed that the
compartment would be a good indicator (e.g.,  Levins  1973, Summers and McKellar 1981).  However,
modeling  approaches are also  limited by lack of data on many wetland species and stressors.  In other
aquatic systems, stable isotope  techniques have been used to collect such data, and their potential for use
in inland wetlands deserves greater attention.

Regardless of the technique used, interpretive caution is  necessary because it is seldom possible to validly
infer trends in all species by monitoring only one or a very few species. Changes in community-level metrics
often give a clearer indication of abnormal biological stress than does the presence or absence of a single
indicator species, regardless of its reputation as being physiologically sensitive or a keystone (Cairns 1974).


                                                 1.3
                     CONSIDERATIONS IN MONITORING PROGRAM DESIGN

Ideally, wetlands should be sampled both before and after a  stressor is introduced.  If the condition of the
wetland prior to stressor introduction is unknown, this  can sometimes be estimated by interpreting historic
aerial photographs or using paleoecological techniques such as seed bank analysis (e.g., Poiani and Johnson
1989), tree ring analysis (Bowers et al. 1985, Hupp and Morris 1990, Sigafoos 1964), sediment core analysis
of pollen  (palynological analysis) (Agbeti and Dickman  1989, Battarbee and Charles 1987) and sediment
accretion (Bloesch and Evans 1982, Ritchie and McHenry 1985).

Also, measurements made in potentially stressed wetlands should be compared with measurements from
reference  wetlands, i.e., structurally similar wetlands that have as many of the following characteristics as
possible:

o       Wetland arose naturally and at a  considerable time  in the past, rather than  being recently
        constructed;

o       Surrounding watershed, particularly within 500 feet of the wetland transition with upland, is largely
        undeveloped;

o       Groundwater flow and  streamflow  to  the wetland  have not  been altered  by withdrawals or
        channelization within about one mile of the wetland;

o       Water  levels  fluctuate naturally, not being affected by diversions, dams, or nearby wells;

o       Wetland has  not been  recently used for silviculture, grazing, or other human uses that potentially
        impact vegetation and/or water quality and quantity.

Efficient wetland biomonitoring also requires knowledge of (a) life history aspects of wetland organisms, (b)

-------
physiology and relative sensitivities to stressors of the component organisms, and (c) dynamics of physical
and chemical factors that largely determine stressor availability.

The optimal seasonal timing from a biological perspective may not coincide with the best timing from a
perspective of physical access. Time-of-day is also an important consideration, particularly when monitoring
vertebrates and in communities having a large component of highly mobile or seasonally secretive species.
The ideal time for sampling is also determined by factors such as the following:

o       Times at which organisms of concern are most likely to be at maximum numbers or detectability,

o       Times at which species composition is most representative of total annual species composition;

o       Times at which organisms are most physiologically sensitive to a particular stressor of interest;

o       Times at which concentration of, or organism exposure to, the stressor is greatest.

Previous studies can be reviewed to indicate the best sampling  time with regard to balancing these factors.
Most biological surveys of wetlands have been conducted during the growing season, and relatively little is
known of exposure or community structure and function during stressful conditions of ice cover,  severe
anoxia, or drought.

Because of the large spatial and temporal variability of wetland environments, sample collections should be
replicated, both within and among wetlands,  and within and among sampling times.   Options for  spatial
arrangement of replicates or multiple samples include the following:

o       Random placement;

o       Along transects (usually perpendicular to wetland gradient or flow and extending to the  deepest part
        of the wetland, and  sometimes intentionally aligned to intersect all  habitat or topographic  "types"
        within the wetland);

o       At ecotones (spatial boundaries between major vegetation types, and open water and  vegetation);

o       In proportion to measured occurrence of habitat types (or hydroperiod classes) present within the
        wetland;

o       At locations subjectively felt by the investigator to represent the wetland.

Statistical protocols are available for estimating  requisite number of samples in wetlands, given a desired
detection level and initial information on sample variability (e.g., Downing and Anderson 1985, Eberhardt
1978, Jackson  and Resh 1988, Resh and Price 1984).


                                                1.4
                 CONSIDERATIONS IN DATA ANALYSIS AND INTERPRETATION

The selection and interpretation of appropriate metrics (variables) is at least as important as the selection
of appropriate taxa and sampling techniques discussed above.   Thus, if  data  are  to  be convened to
information, questions  such as the following must be addressed:

o       Which metric - abundance, biomass, or species richness - is the most sensitive indicator of wetland
        biological change?

-------
o       When are guilds an appropriate way to compile data?

o       Do similarity indices and ordination procedures indicate stress from contaminants better than they
        show stress from hydroperiod alteration?

o       When metrics describing ecosystem structure (such as the above) show that a wetland has changed,
        what can be inferred about the wetland's change in function?

A host of metrics and procedures are available for representing ecological change (Table 3).  To optimize
detection  of ecologically degraded condition, it  is usually best  to  use several of these in combination
(Schindler 1987), as is done by the Index of Biotic Integrity (IBI) that was developed for other surface waters
(Karr 1981).  For situations where only a few metrics can be used, consider the following ranking of metrics
and procedures, listed in descending order of relative sensitivity as inferred  from existing literature by the
review of  Adamus and Brandt (1990):

1. Clustering and Ordination Procedures
2. Similarity Indices
3. Number of Species (per unit area or per  unit effort)
4. Diversity Indices, Biomass, Abundance

Adamus and Brandt (1990)  cautioned that a ranking different than the above might result, depending on
factors such as  the following:

o       Statistical properties of the data set, relative to mathematical characteristics of the metric/procedure;

o       Particular combination of taxa contained in the data set (and associated life histories varying from
        sample to sample);

o       Taxonomic level-of-identiftcation;

o       Wetland or community type;

o       Type of stressor;

o       Spatial scale of measurement;

o       Temporal scale of measurement (e.g., frequency of sampling, time  elapsed since  the stressor was
        maximal);

o       Sampling equipment, level-of-effort, and  techniques used  for collecting the data.

If used alone, a single number from a metric provides little useful information.  The particular taxonomic
composition that led to a summary metric value is often more instructive.  Shifts in taxonomic composition
in response to  contaminants frequently seem likelier to occur than  changes in  richness (total number of
species) or biomass.

Where data on  sensitivities and life histories  of organisms are available, aggregating species-level monitoring
data by functional groups (sometimes called guilds) of  species can provide for more meaningful  data
interpretations.  Factors such as  the following may be used to place a species into a functional group:

o       Trophic level, assumed niche breadth

-------
o       Size, biomass, caloric content
o       Resident vs. migrant status
o       Home range size or dispersal characteristics
o       Geographic distributional (range) response to changing population density
o       Toxicological sensitivity
o       Life history
o       Habitat preference
o       Behavioral characteristics
o       Phenological characteristics
o       Sensitivity to human presence
o       Status as an exotic or indigenous species
o       Harvested vs. protected status

Species richness is frequently correlated with biomass in wetland macrophyte communities (e.g., Nilsson and
Keddy 1988).  However, this is not true in some other wetland biological communities, such as fish  (e.g.,
Tonn 1985).  Moreover, predicting which species  or  functional groups will become dominant following a
wetland disturbance is  generally more difficult than predicting that species composition, overall richness, or
biomass-abundance will change.

All of the commonly-used  metrics and procedures, except for biomass and abundance, commonly employ
species-level data.  Such data are easily collected for taxa such as birds, but are much more difficult to
acquire for microbial  communities, which have large numbers of species, and for which comprehensive
regional references on  taxonomy are virtually nonexistent. The  need for species-level identifications for the
determination of anthropogenic effects is asserted by some studies and disputed by others; the need may
depend on biases of particular metrics, as well as on costs of making more-detailed identifications vs.  costs
of collecting a larger number of samples that are only identified at a  general taxonomic level.

The utility of some metrics and procedures, as well  as  their sensitivity,  may vary by wetland type.  For
example, metrics and procedures that depend on species-level data (richness, ordination, similarity indices)
are obviously less effective in describing the ecological condition of wetlands that characteristically have low
species richness (e.g., breeding bird richness in salt marshes, fish richness in montane wetlands).

Once community metrics have been calculated, the next logical question is "What represents normal (or
desirable) conditions?"  Normal can be defined either in  terms of (a)  the condition of a reference wetland,
(b) average regional conditions, or (c) ecological conditions necessary for sustaining the natural variability
or trends within an ecosystem type and/or its desired functions.  The definition of normal condition should
encompass not only a  mean condition, but the naturally-occurring extremes in structure and function that
may be expected over decades of time (i.e., temporal  and spatial variability).

Caution  always must  be exercised  in interpreting  community-level data as  a  potential  indication  of
anthropogenic stress.  Absence of a species in a wetland  may be due merely to factors not apparent at the
time of sampling.  Sampling metrics,  particularly species richness, are often very sensitive to the intensity
of sampling, i.e., number of samples, level of effort, size and natural heterogeneity of the wetland sampled.
Genetic mutation, natural selection, or adaptation can result in evolution of tolerant ecotypes - local forms
of a species  that  have become tolerant of contaminants.  This can alter competitive relationships and
ultimately, community structure.  Although it is uncertain as to how widespread this phenomenon may be,
it can be locally important and has been documented to occur in communities  of microbes (Baath 1989),
macrophytes (e.g., Christy and Sharitz 1980, McNaughton  et al. 1974), aquatic invertebrates (e.g., Krantzberg
and Stokes 1989, Kraus and Kraus 1986), and amphibians (e.g., Karns 1984).  The possibility that mobile
fish or wildlife are avoiding contaminated areas also should be considered when  evaluating community-
level vertebrate data.  Conversely, wide-ranging biological indicators may not occur even in the "healthiest"
wetlands if most other surrounding wetlands have  been contaminated or altered.

-------
Finally, wetland function should not always be assumed to change whenever the structure of the biological
community changes.  Changes in community composition can sometimes be compensatory, such that new
species  essentially replace the  function of original species and overall community biomass and perhaps
richness does  not change (Cairns  and Pratt 1986, Herricks and Cairns 1982).  Thus, whenever possible
monitoring should address both the structure and function of wetlands.

                                                1.5
               BIRD COMMUNITIES AS AN EXAMPLE OF WETLAND VARIABILITY

Analyses of existing databases can  be  used to document spatial and temporal variability of wetlands. The
USEPA is using  such analyses as one means of  developing efficient  sampling designs for EMAP, and
establishing a  context for interpretation of data that will be collected by EMAP.  Questions of particular
interest include the following:

o       How many replicates should be collected to estimate precisely the richness and density at a single
        time and  station within a wetland?

o       If only one sample can be collected on a given date, how precisely would it represent richness/density
        in other samples collected  on  the same date in other wetlands or different stations within the same
        wetland?

o       If only one sample can be  collected in a wetland, how precisely would it represent richness/density
        in other samples collected in  the same wetland on a different date?

Answers to these  types of questions are likely to vary by region, wetland type, and desired statistical power-
of-detection.

Two of the databases being examined by  the planning group for EMAP-Wetlands are the Breeding Bird
Survey (BBS)  database and the Breeding Bird Censuses (BBC) database. These automated databases were
obtained from the U.S. Fish and Wildlife Service  (USFWS) and the Cornell Laboratory of Ornithology,
respectively.

The Breeding  Bird Survey

The BBS was established in 1966, and covers all 50 states and some Canadian provinces.  Data on bird
relative abundance on a single date during the nesting season have been collected, usually recurrently, from
about 2500 transects ("routes"), each randomly placed within an ecoregion.   Each route is 40 km in length
and contains 50 evenly-spaced data collection points.

One objective of EPA's exploratory analysis for EMAP was to  identify regions in which birds specifically
known  to inhabit wetlands are showing the strongest reductions in distribution.  The severest declines in
nesting birds appear to be occurring in riparian systems of the Great Basin, Prairie Pothole and Cornbelt
areas  of the  Midwest, bottomland hardwood wetlands of the Lower Mississippi  River Valley, southern
Florida, and the Adirondack Mountains of New York (Figure 2). This map, produced by EPA synthesis and
plotting of USFWS data, is based on regional calculations of the average for all wetland species.  For each
wetland species in each region, the ratio of number of routes in which the species is  declining to number
of routes in which it is increasing was calculated.  After ratio values had been  calculated for all wetland
species, the values were averaged  among  all wetland species in  a region, and quantiles were assigned to
regions based on  their average.  The USFWS calculated the trends using the method of Geissler (1984), and
delimited regional boundaries partly based on overall homogeneity in bird community composition.
                                                 8

-------
Each species trend calculated by USFWS has associated with it a variance estimate. For example, a species
with a monotonous, unidirectional trend would have a small trend variance.  The trend variance estimates
of the component wetland species in each region were averaged among species and expressed as regional
quantiles.  This suggested that the largest year-to-year variability in wetland bird densities occurs in the
Great Basin, the Central Valley of California, the Sonoran Desert, east Texas, and southern Florida.  It is
not possible to determine whether this variability is due to variable extent of observer participation in the
BBS, to climatic variability, or other factors.

In reviewing results from  the BBS, several cautions are appropriate.   Density of coverage is geographically
inconsistent, varying from 1 to 16 routes per degree (latitude-longitude) block.  The survey routes are not
located to intentionally intersect wetlands, so wetlands are included opportunistically. Because routes follow
roads and rely largely on auditory detection more suitable for forest birds, they almost surely underestimate
wetland species.  Routes are run only once annually by a single observer, so many species may be missed.
Some routes are conducted later in the season than is optimal  for detecting some wetland species.  Regions
that show no decline in wetland species as a whole may still be experiencing declines of particular wetland
bird species or guilds, or  of wetland species as a  whole  in  some  wetland  types  but  not in others.
Nonetheless, the BBS database, by its sheer quantity of spatial  and temporal coverage, represents a valuable
resource for helping define "average" bird densities (in relative terms) and for aiding detection of regional
trends in wetland birds.

The Breeding Bird Census

Estimates of species richness and density (number of breeding pairs per km2) from wetland Breeding Bird
Censuses more accurately represent breeding density of particular sites than do data from the BBS, because
they are based on repeated visits throughout the nesting season to a specific plot.  However, they represent
one-tenth the number of sampling points, and unlike the BBS,  the census areas do not represent a statistical
probability sample of any region or state.  Thus, comparisons of results among states  or habitats must be
not be considered definitive.

One objective for the analysis of the BBC data was to examine differences  in breeding bird richness and
density among wetland types.  These differences were not tested statistically, but  are presented in Tables 4
and 5. Riparian wetlands had the densest concentrations and arctic wetlands, tidal marshes, and bogs had
the sparsest. When data were grouped by region, some exceptions  were noted.  Shrub wetlands had the
largest densities among wetland types in the Northeast, Prairie  Potholes,  Rockies, Southwest, and California.
Marshes had the largest densities among types in the Ohio-Indiana-Illinois region.  The two greatest densities
of all wetland counts were from riparian willow woodlands in  California, one with 4547 pairs per km^ and
35 species, and the other with 3208 pairs per km^ and 13 species. Other large breeding bird densities were
in a California  lacustrine marsh (3684 pairs, mainly Tricolored Blackbird), and in a cattail bulrush wetland
in North Dakota (3418 pairs, mainly Yellow-headed Blackbird).  By state, the median density of breeding
birds in wetlands ranged from 138 in Alaska to 1857 in North Carolina.

When  species richness rather than density is used as a metric,  habitat rankings differed.  Mixed habitats
(interspersed uplands and wetlands) had, as expected, the most species; forested wetlands (both riverine and
nonriverine) similarly had great richness. Tidal marshes had the fewest species, and fresh marshes were also
relatively species-poor during the nesting season. Again, there  were regional exceptions. Mixed habitats had
fewer species than bogs in the Northeast and fewer species than riparian and shrub wetlands in the East-
Central states.  In contrast, in the Southeast, shrub wetlands were nearly as impoverished as tidal marshes.
The greatest richness (i.e., number of all breeding species per census  plot) recorded in any wetland census
was in a bulrush-cattail marsh in Montana, where 68 species were reported.

A second objective with the BBC data was to examine differences among  wetland types with regard to annual
variability.  Overall, most censuses  had a between-year variation  in  bird  density, as  expressed by the

-------
coefficient of variation, of less than 40%, with a median of about 20% (Figure 3).  Wetland bird species
richness varied somewhat less, with a between-year coefficient of variation of about 14%, and usually less
than 25%. Year-to-year variability in both species richness and nesting density was generally greater in arctic
and emergent wetlands than in woody or riparian types.  This pattern was particularly true when some of
the data were grouped by region.  Annual variability of both density and richness was largest in wetlands
of California,  the southern Rockies, and the North-Central States.   Texas wetlands evidenced low annual
variability in bird density, but  had  large variability in species richness, as did Montana, Virginia, and Maine
to a lesser  extent.  Overall analysis of the 478  BBC  census plots  from wetlands showed  the  following
statistically significant  (p<0.05), linear relationships, based on log-transformed data:

o       Median number  of species was correlated with pair density and  number of repeat censuses (years)
        on a plot;

o       Variability in number of species was inversely correlated with number of  species;

o       Median pair density was not correlated with number of repeat censuses (years) on a plot;

o       Variability in pair density was correlated with  pair density and  number of repeat censuses (years)
        conducted on a plot;

o       Variability in pair  density was correlated with variability in number of species.

However,  there was considerable  scatter in  all of these relationships,  and  the Spearman  nonparametric
correlation coefficients (r) never exceeded 0.5. Analysis  based on regional and wetland-type groupings of
data was not attempted.

The BBC data must be viewed cautiously. Habitat heterogeneity is  not standardized among the censuses,
and the acreage of censused plots is not  consistent among censuses. Thus, because of uncertainty in the
species-area relationships, richness data in particular are  difficult to  compare.  Also, in most cases; census
plots are too small and heterogeneous to adequately census species with large home ranges (Terborgh 1989),
as is typical  in wetlands.    Finally,  because the habitat-based naming of  individual  censuses is not
standardized,  considerable judgement had to be exercised  in  identifying which censuses  were wetland
censuses, and more specifically, to which of the  nine wetland types  listed in Tables 4 and 5 a particular
wetland census should be assigned.

                                                 1.6
                                             SUMMARY

Efforts to develop and compare indicators on wetland ecological condition should  employ  designs that span
a gradient of disturbed and undisturbed (but otherwise as similar as possible) wetlands.  As resources allow,
they should compare all taxa and ecosystem processes, as  well  as metrics and data  reduction techniques,
which from  a theoretical perspective and studies to date show promise for use. They should be regionally-
based, covering specific wetland types  as defined  by predominant hydrologic regime, chemical regime, and
vegetation form. Empirical results should be integrated with results from experiments and simulation models
to identify wetland  components most suitable as  indicators.

                                                 1.7
                                      ACKNOWLEDGEMENTS

Support for activities related to this paper was provided by the. Wetlands component of the USEPA's EMAP
program, and by the USEPA Office of Policy, Planning,  and Evaluation. The work was  performed under
contract 68-C8-006 to NSI Technology Services Corporation. This chapter has been subjected to EPA's peer


                                                 10

-------
review procedures and approved for publication.

                                                1.8
                                       LITERATURE CITED

Adamus, P.R. and K. Brandt. 1990. Impacts on Quality of Inland Wetlands of the United States: A Survey
of Indicators, Techniques, and Applications of Community Level Biomonitoring Data.  EPA/600/3-90/073.
U.S. Environmental Protection  Agency, Cincinnati, Ohio.

Agbeti, M. and M. Dickman. 1989. Use of lake fossil diatom assemblages to determine historical changes
in trophic status.  Canadian Journal of of Fisheries and Aquatic Sciences 46:1013-1021.

Aust, W.M., S.F. Mader, and R. Lea.  1988. Abiotic changes of a tupelo-cypress swamp following helicopter
and rubber-tired skidder timber harvest. Proceedings of the Fifth  Southern Silviculture Research Conference,
Memphis, Tennessee.  National Council  of the Paper Industry for Air and Stream Improvement,  Inc.,
Corvallis, Oregon.

Baath, E. 1989. Effects of heavy metals in soil on microbial processes and populations  (a review). Water,
Air,  Soil Pollution 47:335-379.

Battarbee, R.W. and D.F. Charles. 1987. The use of diatom assemblages in lake sediments as a means of
assessing the timing, trends, and causes of lake acidification.  Progress in Physical Geography 11:552-580.

Beals, E.W.  1973.  Ordination: mathematical  elegance and ecological naivete. Journal of Ecology 61:23-
36.

Bianchi, T.S.  and S. Findlay. 1990. Plant pigments as tracers of emergent and submergent macrophytes from
the Hudson River.  Canadian Journal of Fisheries and  Aquatic  Science 47:492-494.

Bloesch, J. and  R.D. Evans. 1982.  Lead-210 dating of sediments with accumulation  rates estimated by
natural  markers  and  measured with sediment traps.  Hydrobiologia 92:579-586.

Bowers, L.J., J.G.  Gosselink, W.H. Patrick, Jr., and E.T. Choong. 1985.   Influence of climatic trends on
wetland studies in the eastern United States which utilize tree ring data.  Wetlands 5:191-200.

Boyle, T.P., G.M. Smillie, J.C. Anderson, D.R. Beeson.  1990.  A sensitivity analysis of nine diversity and
seven similarity indices. Journal of the Water Pollution Control Federation 62:749-762.

Brooks, R.P., M.J. Croonquist, D.E. Arnold, C.S. Keener, and E.D. Bellis.  1990.  Conservation of Wetland-
Riparian  Ecosystems and Resources:  A  Landscape  Approach.   Final  Report,  Pennsylvannia Game
Commission, Harrisburg, Pennsylvannia.

Cairns,  J., Jr. 1974.  Indicator species vs. the concept of community structure as an index of pollution.
Water Resources Bulletin 10:338-47.

Cairns,  J., Jr. and J.R.  Pratt.   1986.  On the relation between structural and functional analyses of
ecosystems.  Environmental Toxicology and Chemistry 5:785-786.

Canfield, D.E., K.A. Langeland, M.J.  Maccina, W.T. Haller, and J.V. Shireman.  1983.  Trophic state
classification  of lakes with  aquatic macrophytes.  Canadian Journal of Fisheries and Aquatic  Sciences
40(10):1713-1718.
                                                11

-------
Christy, E.J. and R.R. Sharitz.  1980. Characteristics of three populations of a swamp annual under different
temperature regimes. Ecology 6:454-460.

Downing, J.A.  and M.R. Anderson.   1985.  Estimating the standing biomass of aquatic macrophytes.
Canadian Journal of Fisheries and Aquatic Sciences 42:1860-1869.

Eberhardt, L.L.  1978.   Appraising variability in population studies.   Journal of Wildlife Management
42:207-238.

Geissler, P.H.  1984. Estimation of animal population  trends and annual indices from a survey of call-
counts or other indicators.  Proceedings of the American Statistical Association, Sction on Survey Resource
Methods 1984:472-477.

Hellawell, J.M.  1984.   Biological  Indicators of Freshwater Pollution and Environmental Management.
Elsevier Applied Science Publishers, London and New York.  528 pp.

Herricks, E.E. and J. Cairns, Jr.  1982.  Biological Monitoring.  Part III -  Receiving system methodology
based on community structure.  Water Research 16:141-153.

Huhta, V.  1979.  The use of similarity indices for measuring succession in  invertebrate communities, pp.
100-103  In: The  Use  of Ecological  Variables in  Environmental  Monitoring.   The National Swedish
Environment Protection Board, Report PM 1151.

Hupp, C.R. and E.E. Morris.  1990. A dendrogeomorphic approach to measurement of sedimentation in
a forested wetland, Black Swamp, Arkansas.  Wetlands 10:107-124.

Jackson, J.K. and V.H. Resh.  1988. Sequential decision plans in monitoring benthic macroinvertebrates:
Cost savings, classification accuracy, and development of plans.  Canadian Journal of Fisheries and Aquatic
Sciences 45:280-286.

Karns, D.R.  1984.   The relationship of amphibians and reptiles to peatland habitats in Minnesota. Final
Report to Peat Program.  Minnesota Department of Natural Resources. 84 pp.

Karr, J.R.  1981.  Assessment of biotic integrity using fish communities.  Fisheries 6(6):21-27.

Kelly, J.R. and M.A, Harwell.  1989.   Indicators of ecosystem response and recovery,  pp. 9-35 In: S.A.
Levin, M.A. Harwell, J.R. Kelly, and K.D. Kimball.  Ecotoxicology: Problems and Approaches.  Springer-
Verlag, New York, New York.

Krantzberg, G.  and P.M. Stokes.  1989.  Metal regulation, tolerance, and body burdens in the larvae of the
genus Chironomus.  Canadian Journal of Fisheries and Aquatic Sciences 46:389-398.

Kraus, M.L. and D.B. Kraus.   1986.  Differences in the  effect of mercury on predator avoidance in two
populations of  the Grass Shrimp.  Marine Environmental Research 18:277-289.

Landres, P.B., J. Verner, and J.W. Thomas. 1988.  Ecological uses of vertebrate indicator species: A critique.
Conservation Biology 2:316-328.

Leibowitz, N.C., L. Squires, and J.P. Baker.   1991.  Environmental Monitoring and Assessment Program:
Research Plan for Monitoring Wetland Ecosystems.  US Environmental Protection Agency, Environmental
Research Laboratory, Corvallis, Oregon.
                                                12

-------
Levins, S.  1973.  The qualitative analysis of partially-specified systems.   Annals New York Academy of
Science 231:123-138.

McNaughton, S.J., T.C. Folsom, T. Lee, F. Park, C. Price, D. Roeder, J. Schmitz, and C. Stockwell.  1974.
Heavy metal  tolerance in Tvpha latifolia without the evolution of tolerant races.  Ecology 55(5):1163-1165.

Nilsson, C. and P.A. Keddy.  1988.  Predictability of change in shoreline vegetation in  a hydroelectric
reservoir, northern Sweden. Canadian Journal of Fisheries and Aquatic Sciences 45:1896-1904.

Patterson, N.J. and T.H. Whillans.  1984.  Human interference with natural water level  regimes in  the
context of other cultural stresses on  Great Lakes wetlands,  pp. 209-239 In: Prince, H.H., and F.M. D'ltri
(eds.). Coastal Wetlands.  Lewis Publishers, Inc., Chelsea, Michigan.

Pielou, E.C.  1984. The Interpretation of Ecological Data: A Primer on Classification and Ordination. John
Wiley & Sons, New York.  263 pp.

Pilli, A., D.O. Carle, and B.R. Sheedy. 1989. AQUIRE: AQUatic toxicity Information REtrieval data base.
NTIS EPA/DF/MT-89/031.  PB89-70344.

Poiani, K.A. and W.C. Johnson.  1989. Effect of hydroperiod on seed-bank  composition in semi-permanent
prairie wetlands.  Canadian Journal of Botany 67:856-864.

Ramm, A.E.  1988. The community degradation index: a  new method for deterioration of aquatic habitats.
Water Research 22:293-301.

Resh, V.H. and D.G.  Price. 1984. Sequential sampling:  A cost-effective approach for monitoring benthic
macroinvertebrates in environmental impact assessments.   Environmental Management 8(1):75-80.

Ritchie,  J.C.  and J.R. McHenry. 1985.   A comparison of three  methods for measuring  recent rates of
sediment accumulation.  Water Resources Bulletin 21(1): 99-103.

Schaeffer, D.J., E.E. Herricks, and H.W. Kerster.  1988. Ecosystem health:  I.  Measuring ecosystem health.
Environmental Management 12(4):445-455.

Schindler, D.W.   1987.  Detecting ecosystem responses  to anthropogenic stress.  Canadian Journal of
Fisheries and Aquatic Sciences 44(l):6-25.

Sigafoos, R.S. 1964.  Botanical Evidence of Floods and Flood-plain  Deposition.  Professional Paper 485-
A, U.S. Geological Survey, Reston, Virginia.

Summers, J.K. and H.N. McKellar, Jr.  1981.  A sensitivity analysis of an ecosystem  model of estuarine
carbon flow.  Ecological Modelling 13:283-301.

Terborgh, J.  1989. Where Have All the Birds Gone?  Princeton University Press, Princeton, New Jersey.

Tonn, W.M.  1985. Density compensation in Umbra-Perca fish assemblages of northern Wisconsin Lakes.
Ecology 66(2):415-429.
Washington, H.G.  1984.  Diversity, biotic, and similarity indices: a review with special relevance to aquatic
systems.  Water Research 18:653-694.

Wolda, H.  1981.  Similarity indices,  sample size, and diversity.  Oecologia 50:296-302.
                                                13

-------
Table 1.  Indicators Proposed by EPA for Use by EMAP-Wetlands
Physical:



o       Wetland extent and type diversity



o       Landscape and wetland pattern



o       Hydroperiod



o       Sediment and organic matter accretion



o       Chemical contaminants in sediment, tissues of plants and animals






Biological:



o       Vegetation: species composition, spectral greenness, and % cover



o       Birds: community composition, bioaccumulation



o       Amphibians: community composition, bioaccumulation
                                               14

-------
Table 2. Stressors Potentially Occurring In Wetlands
Enrichment/Eutrophication. Increases in concentration or availability of nitrogen and phosphorus. Typically
associated with fertilizer application, cattle, ineffective wastewater treatment systems, fossil fuel combustion,
and urban  runoff.

Organic Loading and Reduced Dissolved  Oxygen.  Increases in  carbon, to the point where an increased
biological oxygen demand reduces dissolved oxygen in sediments and the water column and increases toxic
gases (e.g., hydrogen sulfide, ammonia). Typically associated with ineffective wastewater treatment systems.

Contaminant Toxicity.  Increases in concentration, availability, or toxicity of metals and synthetic organic
substances.  Typically associated with agriculture (pesticide applications), aquatic weed control, mining, urban
runoff, landfills, hazardous  waste sites, fossil fuel combustion, and wastewater treatment systems.

Acidification.  Increases  in acidity  (decreases in pH).   Typically associated  with mining  and fossil fuel
combustion.

Salinization.  Increases in dissolved salts, particularly chloride, and related parameters such as conductivity
and  alkalinity.  Typically associated with  road salt used for winter ice control,  irrigation return waters,
seawater intrusion (e.g., due to land loss or aquifer exploitation), and  domestic/industrial wastes.

Sedimentation/Burial. Increases  in deposited sediments, resulting in partial or complete burial of organisms
and alteration of substrate.  Typically associated with agriculture, disturbance of stream flow regimes, urban
runoff, ineffective wastewater treatment plants, deposition of dredged or other fill material, and erosion from
mining and construction  sites.

Turbidity/Shade. Reductions in  solar penetration of waters  as a result of blockage by suspended sediments
or overstory vegetation or other  physical obstructions.  Typically associated with agriculture, disturbance of
stream flow regimes, urban runoff,  ineffective wastewater treatment plants, and erosion from mining and
construction sites,  as well  as from natural succession, placement of bridges  and other  structures, and
resuspension by fish (e.g., common  carp) and wind.

Vegetation  Removal.  Defoliation  and possibly reduction  of vegetation  through physical removal, with
concomitant increases in solar radiation.  Typically associated with aquatic weed  control, agricultural and
silvicultural activities, channelization,  bank stabilization, urban development,  defoliation  from  airborne
contaminants and other stressors included in this report, grazing/herbivory (e.g., from muskrat, grass carp,
geese, crayfish, insects), disease,  and fire.

Thermal Alteration.   Long-term changes  (especially increases)  in  temperature of water  or sediment
Typically associated with power plants, other industrial  facilities, and global climate wanning.

Dehydration. Reductions in wetland water levels or increased frequency, duration, or extent of desiccation
of wetland  sediments.  Typically associated with ditching,  channelization of nearby streams, invasion of
wetlands by highly  transpirative  plant species, outlet widening, subsurface  drainage, global climate change,
and ground or surface water withdrawals for agricultural, industrial, or residential use.

Inundation. Increases in wetland water levels or increase in  the frequency, duration, or extent of saturation
of wetland sediments. Typically  associated with impoundment (e.g., for cranberry  or rice cultivation, flood
control, water supply, waterfowl  management) or changes in watershed land use that result  in more runoff


                                                  15

-------
being provided to wetlands.

Fragmentation of Habitat. Increases in the distance between, and reduction in sizes of, patches of suitable
habitat.

Other Human Presence.  Increases in noise, predation from pets, disturbance from visitation, invasion by
aggressive  species  capable  of outcompeting species that normally characterize intact  communities;
electromagnetic, ultraviolet (UV-B), and other radiation.
                                                 16

-------
Table 3. Examples of Analytical Metrics, Indices, and Procedures Used for Wetland Community Studies.


Similarity  (Comparative) Indices.   Metrics  that reflect the number  of species or functional  groups in
common between multiple wetlands or time periods.   May be weighted  by relative abundance, biomass,
taxonomic  dissimilarity, or caloric content of the component  species.   Includes Jaccard coefficient, Bray-
Curtis coefficient, rank coefficients, overlap indices, the "community degradation index" (Ramm 1988), and
others.  Results of sensitivity analyses of several indices are reported by Boyle et al. (1990), Huhta (1979),
Washington (1984), Wolda (1981).

Cluster Analysis and Ordination. Procedures that detect statistical patterns and associations in community
data.  Can be used to hypothesize relationships to a  stressor.  Includes principal components  analysis,
reciprocal averaging, detrended correspondence analysis, TWINSPAN, canonical correlation, and others. Can
be used to identify guilds (see below).   A  useful reference  is Pielou (1984), and a cautionary note is
expressed by  Beals (1973).

Food  Web  Analysis.  Procedures that measure  length of food chains, number of trophic levels,  ratio of
number of  trophic species to trophic links, and similar measures.  As yet, these procedure have been tested
in stressed  wetlands in only a few cases.
                                               >
Tolerance Indices.  Metrics that reflect proportionate composition of tolerant vs. intolerant taxa.  Includes
saprobic indices, macroinvertebrate EPT index,  Hilsenhoff index,  and others detailed  and  compared in
Hellawell (1984) and Washington (1984).  "Tolerance" usually means tolerance to organic pollution; tolerance
to many toxicants and physical habitat alterations may not be  well-reflected by available indices.

Functional Group (Guild) Analysis. Procedures in which individual species are assigned to functional groups
(species assemblages) based on similar facets of their life history, sensitivity, or other factors.

Indices  of Biotic Integrity. Indices that are a composite of weighted metrics describing richness, pollution-
tolerance, trophic levels,  abundance, hybridization, and deformities.  Widely used in stream fish studies (see
Karr 1981).
                                                 17

-------
 TABLE 4.
BREEDING BIRD CENSUS ESTIMATES OF RICHNESS  BY REGION AND WETLAND TYPE
 REGION AND HABITAT TYPE

 ALL REGIONS COMBINED
 Arctic/subarctic
 Bog
 Forested nonriverlne swamp
 Fresh marsh
 Mixed
 Riparian or riverine forested
 Shoreline
 Shrub (riverine or npnriverine)
 Tidal marsh
NORTHEAST:  HJ. NY. CT. RI. HA. VT, NH. HE
Bog
Forested nonriverine swamp
Fresh marsh
Mixed
Riparian or riverine forested
Shoreline
Shrub (riverine or nonriverine
Tidal marsh
MID-ATLANTIC:  PA. MD. DE. VA. UV
Bog
Forested nonriverine swamp
Fresh marsh
Mixed
Riparian or riverine forested
Shrub (riverine or nonriverine)
Tidal marsh
SOUTHEAST:  NC. SC. 6A. FL. AL. MS. TN. KY
Forested nonriverine swamp
Fresh marsh
Riparian or riverine forested
Shore!1ne
Shrub (riverine or nonriverine)
Tidal marsh
CENTRAL:  IL. IN. OH
Bog
Forested nonriverine swamp
Fresh marsh
Mixed
Riparian or riverine forested
Shrub
                                            N'
                                             27
                                             93
                                             87
                                            146
                                            111
                                            321
                                             28
                                             98
                                             44
                                             36
                                             42
                                             19
                                             32
                                             19
                                             11
                                             33
                                             11
                                             11
                                              1
                                              3
                                              7
                                             64
                                              5
                                              8
                                             26
                                              1
                                             34
                                              2
                                              3
                                             10
                                            31
                                             2
                                            23
                                             5
                                             4
                                            15
NUMBER
MED"
11.0
21.0
26.0
9.5
32.0
25.0
18.0
22.0
5.0
29.5
26.5
11.0
30.0
19.0
24.0
22.0
8.0
23.0
19.0
14.0
42.0
26.0
24.0
5.5
26.0
13.0
24.5
5.0
5.0
3.0
16.0
16.5
6.0
34.0
41.0
36.0
OF SPECIES
Ql"
8.00
15.50
20.00
6.00
27.00
20.00
16.25
13.00
3.00
24.00
20.00
6.00
23.50
15.00
12.00
20.50
5.00
22.00
19.00
10.00
32.00
24.00
19.00
2.50
21.25
13.00
21.00
5.00
5.00
3.00
8.00
16.00
3.00
30.50
38.25
16.00
Q3"
14.00
28.00
28.00
20.00
37.00
31.50
24.75
33.00
 9.75
35.75
29.25
22.00
33.75
32.00
25.00
28.00
11.00
27.00
19.00
24.00
44.00
30.75
34.00
27.25
28.00
13.00
28.25
 5.00
 5.00
 3.25
18.00
17.00
 9.00
38.50
46.75
41.00
"Med.= Median.  Ql=25th Quartile. Q3=75th Quartile
                                                 18

-------
 HABITAT TYPE

 NORTH  - CENTRAL:  HN. VI. MI
 Bog
 Forested nonriverine swamp
 Fresh  marsh
 Mixed
 Riparian or riverine forested
 Shoreline
 Shrub  (riverine or nonriverine)
SOUTH - CENTRAL:  AR. LA
Forested nonriverine swamp
Riparian or riverine forested
SOUTHWEST:  TX. NM. OK
Mixed
Riparian or riverine forested
CENTRAL:  IA. MO. HE. KS
Fresh marsh
Riparian or riverine forested
PRAIRIE AND ROCKIES:  SO. NO. MT. CO. UT. VY
Fresh marsh
Mixed
Riparian or riverine forested
Shorel1ne
Shrub (riverine or nonriverine)
WEST:  AZ. NV. CA
Fresh marsh
Riparian or riverine forested
Shoreline
Shrub (riverine or nonriverine)
Tidal marsh
NORTHWEST:  10. OR. WA. AK
Arctic/subarctic
Bog
Riparian or riverine forested
Shrub (riverine or nonriverine)
 6
15
27
40
17
 9
14
 1
20
 7
33
25
 5
33
 3
69
 4
15
15
52
 2
11
15
27
 7
 4
 2
NUMBER OF
HED"
17.5
27.0
13.0
34.0
29.0
18.0
33.0
29.0
27.0
13.0
28.0
8.0
26.0
17.0
32.0
18.0
21.0
15.0
12.0
29.0
25.5
22.0
6.0
11.0
12.0
19.5
13.0
SPECIES
01"
12.75
23.00
8.00
30.25
22.00
17.00
13.75
29.00
23.00
11.00
23.00
5.00
16.50
8.00
21.00
12.50
15.00
11.00
6.00
24.00
18.00
13.00
3.00
8.00
11.00
5.25
13.00
                                Q3"
20.75
30.00
26.00
37.00
32.00
19.00
34.00
29.00
29.75
20.00
41.00
11.00
32.50
53.50
39.00
24.50
24.75
20.00
20.00
36.00
33.00
36.00
 9.00
14.00
15.00
34.50
13.00
"Med.= Median.  Ql=25th Quartlle. Q3=75th Quartlle
                                                 19

-------
 TABLE  5.
BREEDING BIRD CENSUS ESTIMATES  OF DENSITY, BY REGION AND  WETLAND TYPE
REGION AND HABITAT TYPE

ALL REGIONS COMBINED
Arctic/subarctic
Bog
Forested nonriverine swamp
Fresh marsh
Mixed
Riparian or riverine forested
Shoreline
Shrub (riverine or nonriverine)
Tidal marsh
NORTHEAST:  NJ. NY. CT, RI. HA. VT. HH. HE
Bog
Forested nonriverine swamp
Fresh marsh
Mixed
Riparian or riverine forested
Shoreline
Shrub (riverine or nonriverine)
Tidal marsh
MID-ATLANTIC:  PA. HO. DE. VA. WV
Bog
Forested nonriverine swamp
Fresh marsh
Mixed
Riparian or riverine forested
Shrub (riverine or nonriverine)
Tidal marsh
SOUTHEAST:  NC. SC. 6A. FL. AL. MS. TN. KY
Forested nonriverine swamp
Fresh marsh
Riparian or riverine forested
Shore1i ne
Shrub (riverine or nonriverine)
Tidal marsh
NORTH - CENTRAL:  MN. VI. HI
Bog
Forested nonriverine swamp
Fresh marsh
Hixed
Riparian or riverine forested
Shoreline
Shrub (riverine or nonriverine)
                                            DENSITY (Pairs per square kilometer)

                                            N         MED"      Ql"        Q3"
                                            36
                                            42
                                            19
                                            32
                                            19
                                            10
                                            33
                                            11
                                             11
                                              1
                                              3
                                              7
                                             64
                                              5
                                              7
                                             26
                                              1
                                             34
                                              2
                                              3
                                             10
                                             7
                                            15
                                            27
                                            40
                                            17
                                             9
                                            14
138.0
327.5
667.0
667.5
667.0
886.0
872.0
916.5
330.0
554.5
321.0
727.0
808.0
580.0
900.5
1703.0
659.0
601
618
695
1941
1143
818
279
2134.0
96.0
795.0
44.5
220.0
307.0
538
801
874
611
471
1210
998
100.00
269.50
309.00
348.25
578.00
497.50
367.00
575.50
264.00
291.00
251.50
622.00
357.50
272.00
522.50
797.50
385.00
560.00
618.00
469.00
815.00
864.75
603.00
143.00
749.00
96.00
541.00
37.00
49.00
200.75
314.00
741.00
601.00
578.00
391.50
1162.00
854.75
159.00
791.25
1166.00
973.00
987.00
1421.00
1178.00
1824.50
584.00
1428.00
640.75
877.00
1001.25
860.00
955.50
1967.50
824.00
914.00
618.00
878.00
1106.00
1539.25
1196.50
1203.00
2183.75
96.00
1350.75
52.00
321.00
360.00
852.00
1109.00
1136.00
660.75
581.50
1290.50
1077.75
"Hed.= Median.  Ql=25th Quart)le. Q3=75th Quartile
                                                   20

-------
 REGION AND HABITAT TYPE

 EAST - CENTRAL:   IL.  IN.  OH
 Bog
 Forested nonrlverlne  swamp
 Fresh marsh
 Mixed
 Riparian or riverine  forested
 Shrub (riverine  or nonriverine)
 SOUTH - CENTRAL:  AR.  LA
 Forested nonrlverlne swamp
 Riparian or  riverine forested
SOUTHWEST:  TX. NM. OK
Mixed
Riparian or riverine forested
CENTRAL:   IA. HO. HE. KS
Fresh marsh
Riparian or riverine forested
PRAIRIE AND ROCKIES:  SO. ND. MT. CO. UT. VY
Fresh marsh
Mixed
Riparian or riverine forested
Shore1i ne
Shrub (riverine or nonriverine)
WEST:  AZ. NV. CA
Fresh marsh
Mixed
Riparian or riverine forested
Shoreline
Shrub (riverine or nonrlverlne)
Tidal marsh
NORTHWEST:  ID. OR. WA. AK
Arctic/subarctic
Bog
Riparian or riverine forested
DENSITY (Pairs per square kilometer)

N          MED"      Ql"        Q3"
 1
20
 7
33
25
 5
33
 3
69
 4
15
15
17
52
 2
11
15
27
 7
 4
280.0
265.0
950.0
696.0
727.5
361.0
1810.0
945.5
445.0
430.0
257.0
627.0
586.0
694.0
907.0
228.0
1219.0
829.0
1112.0
1120.5
375.5
1570.0
305.0
138.0
145.0
859.5
235.00
215.00
700.00
604.50
473.50
338.00
1810.00
743.50
314.00
326.00
211.00
388.00
372.00
608.00
325.50
68.00
635.00
213.00
806.50
708.50
356.00
1285.00
273.00
100.00
138.00
129.50
450.00
315.00
1220.00
729.00
842.00
648.00
1810.00
1170.00
697.00
912.00
314.00
975.50
726.50
1098.00
1476.50
385.00
1875.00
2909.00
1396.50
1564.50
395.00
2305.00
438.00
159.00
172.00
1986.25
"Med.= Median.  Ql=25th Quartlle. Q3=75th Quartlle
                                                   21

-------
Aeciricy of tit* locitltu
         to ki + »r -
Figure 1.   Some inland wetlands having  biological community measurements
                                 22

-------
                                       Reproduced from
                                       best available copy.
    lilio  of Dterciiiif  to
     licreiiioi Triniecli
 (li|. •( ill tetliod ipeeiei)

Q  lisifficieil d«t«

23  (ore triiKClt iocrtuioj tbi>
    friiiilii| >tltl Kilintl lillillNi. IIH-IIII
 rri)iri< 1) ISIPI  ItlliKi Ititiret ri>|ti>. Iliri tiiinixild lii«iicl III. Citnllli. »r<|"


                          111 IMHMII*! If htl Irlil
Figure  2.    Regions  of wetland  breeding  bird  species  decline
                                                23

-------
    10  20   30  40   50  60   70  80   90  100  110  120  130
                 Coefficient of Variation (I)
Figure 3.  Annual variation in wetland bird density:
           cumulative frequencies of coefficients of variation
           from Breeding  Bird Censuses     _   •
                             24

-------