COMBINING HGM AND EMAP PROCEDURES TO ASSESS WETLANDS AT THE
WATERSHED SCALE ? STATUS OF FLATS AND NON-TIDAL RIVERINE
WETLANDS IN THE NANTICOKE RIVER WATERSHED, DELAWARE AND
MARYLAND (USA)
Dennis F. Whigham1, Amy Deller Jacobs2,3, Donald E. Weller1, Thomas E. Jordan1, Mary E. Kentula4,
Susan F. Jensen5, and Donald L. Stevens, Jr.5
1Smithsonian Environmental Research Center
Box 28
Edgewater, Maryland, USA 24037
E-mail: whighamd@si.edu
2The Nature Conservancy of Delaware
110 West 10th Street, Suite 1107
Wilmington, Delaware, USA 19801
3Present address:
Delaware Department of Natural Resources and Environmental Control
Division of Water Resources
820 Silver Lake Blvd., Suite 220
Dover, Delaware, USA 19904
4U.S. Environmental Protection Agency
National Health and Environmental Effects Laboratory
Western Ecology Division
200 SW 35th Street
Corvallis, Oregon, USA 97333
5Department of Statistics
Oregon State University
Corvallis, Oregon, USA 97331
Abstract: The hydrogeomorphic (HGM) approach to wetland assessment was combined with the
Environmental Monitoring and Assessment Program (EMAP) survey design procedures to evaluate the
condition of non-tidal riverine and flats wetlands in the Nanticoke River watershed (Delaware and
Maryland, USA). We found degradation of wetland functions below reference standard levels for the
majority of wetlands in both classes. Wetland condition was also related to the level of disturbance in
both wetland classes. In flats, the most common disturbances were associated with hydrologic and
vegetation modifications. Flat wetlands with low HGM function scores for the Plant Community and
Habitat functions had almost all been converted from hardwood forest to Loblolly pine plantations.
Most modifications associated with riverine wetlands were associated with stream channelization. Results
of this study demonstrate that a site-specific and reference-based approach to assessment (i.e., the HGM
method) can successfully be applied at the scale of an entire watershed if it is combined with a sampling
approach that allows sites to be selected without geographic bias. The approach can also be used to
determine if wetland functions vary from one sub-basin to another, and results of this project can be used
by managers to begin to develop strategies for restoration of wetland functions at the watershed scale.
Key Words: Chesapeake Bay, GIS, HGM, watershed, wetland assessment
INTRODUCTION
Most methods that have been developed to assess
the ecological condition and functions of wetlands
emphasize individual wetlands (Bartoldus 1999,
Fennessy et al. 2004), and many of them involve
detailed, site-specific assessments (Brooks et al.
2004, Fennessy et al. 2004). Fewer methods have
been developed to assess habitat integrity or wetland
functions at the scale of an entire watershed (e.g.,
WETLANDS, Vol. 27, No. 3, September 2007, pp. 462?478
? 2007, The Society of Wetland Scientists
462
Zampella et al. 1994, Montgomery et al. 1995,
Bedford 1996, Abbruzzese and Leibowitz 1997,
Lemly 1997, Cormier et al. 2000, Detenbeck et al.
2000, Leibowitz et al. 2000, Tiner et al. 2000, Tiner
2004, 2005). Most watershed-scale assessment meth-
ods do not involve detailed, site-specific assessments;
instead, they provide an overview of habitat in-
tegrity or assess potential wetland functions (e.g.,
Tiner et al. 2000, Tiner 2004, 2005), equivalent to
a level-one or landscape assessment as described by
Brooks et al. (2004) and Fennessy et al. (2004).
Sutter et al. (1999) developed a method that uses
spatial (e.g., GIS) data to assess the condition of
individual wetlands in the context of a watershed or
landscape, but we are not aware of any efforts to test
or utilize the method. Lee et al. (2003) developed
HGM models for all waters/wetlands in the Santa
Margarita River watershed (California), but simi-
larly we are unaware of any efforts to apply the
models to determine wetland condition at the
watershed or sub-basin scales in the Santa Marga-
rita.
To make watershed-scale decisions about wetland
resources, information is needed at all levels of
assessment (Thomas and Lamb 2004, Tiner 2004),
but ultimately, decisions that focus on restoration
design require site-level information for individual
wetlands or groups of wetlands. In other words, it
would require site-level information using either
rapid or intensive assessments (Brooks et al. 2004,
Fennessy et al. 2004). In this paper, we describe an
approach to watershed-scale assessment that com-
bines site-level data produced by using the hydro-
geomorphic (HGM) approach, with the U.S.
Environmental Protection Agency?s (EPA) Environ-
mental Monitoring and Assessment Program
(EMAP) procedures for selecting sampling sites.
The project focused on the Nanticoke River
watershed in Maryland and Delaware (Figure 1).
The Nanticoke River watershed is one of the most
biologically important and wetland-rich watersheds
in the mid-Atlantic region (The Nature Conservancy
1994); yet, many of its natural habitats have been
degraded (Tiner et al. 2000, Tiner 2004, 2005). More
than 40% of the watershed is covered by forest, but
about the same amount is in agriculture, and
agriculture and forest management have been
supported by extensive drainage and channelization
(Tiner 1985, 2004, 2005; Tiner and Burke 1995;
Tiner et al. 2000).
The project had three goals. The first goal, the
focus of this paper, was to assess the ecological
condition of non-tidal riverine wetlands and flats,
the two most widespread and abundant non-tidal
classes of wetlands, at the watershed scale. The
second goal, the focus of Weller et al. (2007), was to
determine if analyses of spatially mapped data
typically used for landscape assessments, could be
used in combination with field-based HGM assess-
ments to predict the condition of individual wet-
lands or groups of wetlands at the watershed scale.
The third objective, Jordan et al. (2007), was to
determine if the HGM models, referred to as HGM
functions, that were developed for the project
adequately characterized ecological processes in
wetlands.
METHODS
Procedures used throughout the project followed
U.S. Army Corps of Engineers guidelines for
developing HGM models (http://www.wes.army.
mil/el/wetlands/guidebooks.html). The HGM mod-
els were developed, calibrated, field tested, and
applied by an interdisciplinary team in field and
workshop formats. Collection of reference data,
testing of assessment procedures, and assessment of
sites selected by EMAP procedures were conducted
by trained field teams of paid staff and volunteers.
In addition to details of the methods described
here, information can also be found in Whigham et
al. (2003). In year 1, we selected a preliminary list of
Figure 1. Map of the Nanticoke River watershed
(hatched area) and its relationship to Chesapeake Bay.
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 463
HGM variables, collected reference data for 23 non-
tidal riverine (hereafter referred to as riverine) and
19 flats wetlands, selected the HGM variables
(Table 1), scaled the variables based on data from
reference sites, and finalized the HGM models
(Table 2) used to conduct site assessments in year
2. In year 1, we also developed, field-tested, and
finalized office and field protocols that followed
procedures approved by EPA for the project
(Whigham et al. 2000). Nomenclature for plant
names follows Radford et al. (1968).
Site Selection
Much of the first year was devoted to identifying
and screening potential assessment sites. The first
step was to evaluate existing digital wetland maps to
develop a single sampling frame from which
potential assessment sites could be selected. Three
digital wetland maps were available. The National
Wetland Inventory (NWI) created wetland maps for
the entire watershed from 1981?82 aerial photogra-
phy (http://www.nwi.fws.gov). Subsequent state
mapping efforts updated the NWI classification by
using more recent aerial photography, (1992 for
Delaware (State of Delaware 1994) and 1988?89 for
Maryland (http://dnrweb.dnr.state.md.us/gis/data/
samples/doqqbase.html)), selecting more field-verifi-
cation sites, and using smaller minimum mapping
units to produce maps that included more wetlands,
particularly smaller ones. We used 1996 Delaware
state wetland maps prepared from 1992 aerial
photography (D.S. Wallace, Delaware Department
of Natural Resources and Environmental Control)
for the 63% of the watershed in Delaware. The
Maryland state wetland maps (http://dnrweb.dnr.
state.md.us/gis/data/) were not complete when we
prepared the sampling frame, but we used the
Maryland state data that were available for another
19% of the watershed. For the remaining 18% of the
Table 1. Variables used in HGM models (Table 2) for
flats and non-tidal riverine classes in the Nanticoke River
watershed. Variable scores varied from 1.0?0.1, and there
were two types (continuous and categorical). Continuous
variables are shown as CO and categorical variables as
CA.
Flats Class Type Descriptor
VDISTURB CA Evidence of vegetation disturbance
VDRAIN CO Percent assessment area affected
by drainage
VFILL CA Presence of anthropogenic derived
sediment
VHERB CA Species of herbs present
VMICRO CA Presence of microtopographic
features
VRUBUS CA Presence of Rubus sp.
VSHRUB CO Shrub density
VSNAG CA Density of standing dead trees
VTBS CO Basal area of trees
VTDEN CO Tree density
VTREE CA Tree species composition
Riverine Class Type Descriptor
VFARBUFFER CA Condition of buffer 20?100 m
from wetland
VFLOODPLAIN CA Floodplain conditions
VINVASIVE CO Presence of invasive species
VNEARBUFFER CO Condition of buffer 0?20 m from
wetland
VSAPLING CA Sapling species composition
VSHRUB CO Shrub density
VSTREAMIN CA Stream condition inside
assessment area
VSTREAMOUT CA Stream condition outside
assessment area
VTBA CO Basal area of trees
VTDEN CO Tree density
VTREE CA Tree species composition
VDISTURB CA Vegetation disturbance
Table 2. HGM models used to calculate functional capacity index (FCI) scores for non-tidal riverine and flats wetland
classes in the Nanticoke River watershed. Variables used in the models are listed and described in Table 1.
Flats Functions
Hydrology FCI 5 0.25 * VFILL + 0.75 * VDRAIN
Biogeochemistry FCI 5 (VMICRO + ((VSNAG + VTBA + VTDEN) / 3)) / 2 * Hydrology FCI
Plant Community 5 ((VTREE + VHERB) / 2) * VRUBUS
Habitat 5 (VDISTURB + ((VTBA + VTDEN) / 2) + VSHRUB + VSNAG) / 4
Riverine Functions
Hydrology FCI 5 SQRT (((VSTREAMIN + (2 * VFLOODPLAIN)) / 3) * VSTREAMOUT)
Biogeochemistry FCI 5 VTBA * Hydrology FCI
Plant Community 5 (0.75 * ((VTREE + VSAPLING) / 2)) + (0.25 * VINVASIVE)
Habitat 5 ((((VTBA + VTDEN) / 2) + VSHRUB + VDISTRUB) / 3 + VSTREAMIN) / 2
Landscape 5 (0.5 * VNEARBUFFER) + (0.25 * VFARBUFFER) + (0.25 * VSTREAMOUT)
464 WETLANDS, Volume 27, No. 3, 2007
Nanticoke watershed that was in Maryland, we used
the NWI wetland maps. The three data sources were
merged into a single digital map layer using ESRI
ArcINFO geographic information system software.
A sample of potential assessment sites was
selected using the EMAP approach for drawing
a sample. More information on the EMAP ap-
proach to sample design and how it was applied to
wetlands can be found in Stevens and Hornsby
(2007). In this case, a random tessellation stratified
(RTS) sample (Dalenius et al. 1961, Overton and
Stehman 1993, Stevens 1997) was drawn from the
Nanticoke watershed sampling frame. An RTS
sample is selected by randomly placing a regular
grid over the target region and then drawing one
point at random within each grid cell. Points that
did not fall in a wetland polygon then were
discarded. Much of the watershed is privately
owned, and some difficulty in getting access
permission to sample on private property was
anticipated (Lesser 2001). To allow for lack of
access, the initial draw consisted of 1,992 points, or
approximately 10 times the target sample size. These
points and their associated map coordinates were
arranged in replicate groups of 25 points each, where
each group in itself was a spatially balanced random
sample of the wetland polygons. Hierarchical
randomization (Stevens and Olsen 1999, 2000) was
used to define the groups, and the groups were
evaluated sequentially. For the sequential evalua-
tion, an initial set of groups was selected, and the
potential sampling sites were evaluated to ensure
that the points were in the correct wetland class and
that access permission could be obtained. If the
initial set does not yield a sufficient sample size of
the desired wetland classes, then the next group in
sequence is evaluated, and so on, until a sufficient
sample size is obtained. This procedure results in
a spatially well-balanced probability sample of the
accessible wetland population.
We evaluated 1,050 of 1,992 potential sites
generated by RTS. Fifty-eight percent (604) of the
1,050 sites were eliminated from further consider-
ation based on the following procedures: 1) sites
were found to be in another wetland class or were
not wetlands based on examination of digital
orthophotographs, wetland maps, USGS topo-
graphic maps, and county soil surveys; 2) sites were
eliminated because they were chosen for testing
HGM models; and 3) a sizable number of flats were
eliminated once the selection process resulted in
a sufficient number of sites in that class, and we only
needed additional riverine sites, which are compar-
atively less common, for the sample. We eventually
selected 446 potential sites (211 riverine and 235
flats) for which we attempted to gain permission to
access. First, we determined ownership and contact
information for landowners. In Maryland, this
information was available in digital form for all
counties through the Maryland Property View CD-
ROM disks (Maryland Property View, 2002 Edition,
Provided by Maryland Department of Planning,
Planning Services Division, Baltimore Maryland,
Copyright Maryland Department of Planning). In
Sussex and Kent County, Delaware, landowner
contact information was obtained from hardcopy
maps and associated tax records. Initial landowner
contact occurred via letter. Written contact included
a cover letter providing a brief introduction to
project goals, scope, and anticipated benefits and
a request for permission to access the wetland area
of interest. A project brochure was included as an
attachment. If phone numbers were available, letters
were followed with a phone call to the landowners.
If phone numbers were unavailable, access permis-
sion was based on positive returns of reply forms
that were included with the cover letter.
We obtained access to all publicly owned sites
(48). Of the privately owned sites (398) that we
attempted to access, we were unable to contact
landowners for 42% of the sites; however, when we
did make contact with landowners, we had a success
rate for obtaining access of 67%. Accessible sites
were field-checked to verify that they were in fact
wetlands (based on the presence of wetland plant
species and soil or water indicators that clearly
showed that wetland conditions were present) and to
categorize them as either riverine or flats. Wetlands
that were not in either class were deleted from the
list of potential assessment sites. The screening
process eventually resulted in assessment of 54
riverine and 89 flats (Figure 2) and took 145
person-days (1,160 hours).
As stated above, we encountered a sizable number
of privately owned sites for which access permission
was denied or was neither explicitly given nor denied
(hereafter referred to as non-response sites). The fact
that we were unable to sample these sites raised
concerns about the representativeness of the
achieved sample and the validity of extrapolating
from the achieved sample to the entire watershed.
The extension of results from the accessible sub-
population to the entire watershed is possible only
with the additional assumption that ??the non-
response is determined by a random selection
mechanism unrelated to the response,?? (i.e., the
responses are missing completely at random
(MCAR)). However, we were more likely to get
permission to access publicly owned sites than sites
that were privately owned. As a result, the sample
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 465
sites that were accessible had a disproportionate
number of publicly owned sites (Pearson?s x2 5 62.6,
df 5 1, p 5 0.003). These results raised the concern
that the MCAR assumption was not tenable because
differing management practices between publicly
and privately owned wetlands could affect wetland
condition and function. Therefore, we corrected for
this sample bias by post-stratifying on ownership.
Details of the post-stratification adjustment are in
Stevens and Hornsby (2007).
Field Procedures
Procedures used to collect data in the field are
described here and in Whigham et al. (2000).
Sampling points were located by using a geographic
positioning system (GPS), in which the coordinates
of each sample point had been entered. Once the
sampling point was located, an assessment area
(AA) was established by centering a 40-m-radius
circle (0.5 ha) around the point. Information on
alterations within the AA was recorded on a stan-
dard form, by category of alteration. Hydrologic
condition was assessed by mapping the location of
ditches, measuring their depth, and estimating how
far each extended outside the AA up to a maximum
of 300 m. Borings were made throughout the AA to
determine the dominant soil series. A nested-plot
design was used to characterize vegetation. Three
randomly placed 7.98-m-radius circular plots (0.02-
ha) were established. Within each of these plots,
evidence of recent logging activities was recorded,
the diameter at breast height of all trees $ 15 cm
was measured, and downed logs, snags, and saplings
were counted. A 2.25-m-radius plot was established
in the center of each 0.02-ha plot, and the number of
stems or clumps of each species of shrub and the
number of tree seedlings were counted. Presence
(i.e., one or more rooted plants) of blackberry
species, (e.g., Rubus allegheniensis Porter) was
recorded in the 2.25-m-radius plots. The presence
or absence of Rubus was chosen as an indicator
variable because blackberry species were common in
all altered sites in the reference data set. Four 1-m2
plots in each 0.02-ha plot (12 per site) were used to
estimate cover of all non-vine herbaceous species,
Sphagnum spp., Mitchella repens L., and unvege-
tated ground. The plots were placed at the end of
each of two transects that perpendicularly bisected
the tree plot and had their origin at the center of the
plot. Mitchella was included as a field indicator in
the herb plots and used to score the VHERB variable
because analysis of reference data demonstrated that
it was common in sites that had not been altered.
Riverine wetlands also were sampled by locating
coordinates with a GPS unit and establishing an AA
by centering a 1-ha area around the point. The area
sampled was 100 m by 100 m, but in a few instances,
adjustments were made when the distance between
the stream channel and the upland was less than
100 m. When those circumstances occurred, the
plots were 200 m by 50 m. Information on human
alterations and qualitative assessment of floodplain
condition within the AA was recorded on a standard
form. Hydrology was assessed inside the AA by
walking the area along the stream channel and
noting alterations such as fill, channelization, and
levees. The same information was recorded along
a 500-m stretch of the channel upstream and
downstream from the edge of the AA. Vegetation
was characterized using the same nested-plot design
Figure 2. Distribution of wetland assessment sites in the
non-tidal portion of the Nanticoke River watershed. The
locations of the three major sub-basins within the
watershed are also shown.
466 WETLANDS, Volume 27, No. 3, 2007
used for flats. The condition of the buffer associated
with the floodplain was documented by recording
the dominant land use in a 100-m by 100-m area that
began at the toe slope of the floodplain on either
side of the AA and extended into the upland. Each
area was divided into a grid of 10 m 3 10 m cells,
and the dominant land use in each cell was recorded.
For all sites, a Disturbance Index (DI) was
calculated from information recorded on the field
data sheets. For all categories of entries on the data
sheets, we determined which ones indicated some
level of human alteration (e.g., presence of ditching
in the AA, presence of fill in the AA, evidence of
logging in the AA). We then calculated the DI for
each site as a percentage of the maximum number of
alteration categories. A listing of the alteration
categories used to calculate the DI for each wetland
class is provided in Table 3.
Data Analysis
Variable Scores were entered into spreadsheets,
and the equations for the HGM models (Table 2)
were used to calculate Functional Capacity Index
(FCI) scores. The distribution of FCI scores was
examined by generating cumulative distribution
(CDF) plots for the flats and riverine populations
(Whittier et al. 2002). Besides showing the distribu-
tion of assessment scores over the sample popula-
tion, CDF plots allow one to estimate what percent
of the wetland area of the population is less than or
equal to a particular FCI score. In this way, one can
see how far the population departs from reference
standard condition or some FCI score of interest.
The relationship of the sites to each other was also
examined by applying a variety of non-parametric
procedures, and it was determined that the relation-
ships among sites and functions was best described
by using the FCI scores in a Non-metric Multi-
dimensional Scaling (NMS) ordination (McCune
and Meffort 1999). Sorensen?s similarity index was
used as the distance measure in the ordinations, and
based on preliminary exploration of the data, the
analyses were run with two axes. For both classes,
more than 95% of the variance was explained by the
two axes. For the riverine class, a final solution to
the ordination, based on procedures described in the
manual (McCune and Meffort 1999), was reached
after 26 iterations and the final stress was 5.1. For
the flats class, two separate ordinations were run
(the reasons are described in detail in the Results
section). For flats that had not been altered in the
past 50 years (determined at each site by the field
assessment team based on professional judgment,
coring trees to determine age, and information from
landowners), the final solution of the ordination was
reached after 37 iterations and the final stress was
9.7. A final solution was reached after 37 iterations,
for flats altered in the past 50 years. The final stress
for this group of sites was 7.2.
To examine spatial variation in wetland condition
within the Nanticoke River watershed, we examined
FCI scores in three major sub-basins of the
watershed: Marshyhope Creek, Broad Creek, and
main-stem Nanticoke River (Figure 2). We used
state watershed boundaries (ftp://dnrftp.dnr.state.
md.us/public/SpatialData/Watershed/WshedBndry/
swshed12.htm, http://www.dnrec.state.de.us/dnreceis/
downloads/zip/watersheds.zip) to identify assess-
ment points within each sub-basin, then applied
analysis of variance (Proc GLM, SAS 2004) to test
for differences among the sub-basins in the FCI
scores for both wetland classes. When we found
significant sub-basin effects on an FCI score, we
compared the mean score for the three sub-basins
(REGWQ, SAS 2004) to identify which sub-basins
were significantly different.
RESULTS
Condition of Non-Tidal Riverine and Flats
Wetlands at the Watershed Scale
Variable scores and FCI scores for each site are
provided in Appendix I for riverine sites and
Appendix II for flats sites. We use several ap-
proaches to evaluate and describe the condition of
wetlands in the two classes. First, the condition is
Table 3. Disturbance categories for each HGM wetland subclass (non-tidal riverine, flats). The Disturbance Index (DI)
was calculated as the average number of disturbance categories that were present on a site as recorded on field data sheets.
Disturbance Category Riverine Flats
Evidence of physical site disturbance, other than ditching and channelization X X
Evidence of vegetation disturbance (e.g., logging) X X
Ditches and levees present in assessment area X X
Evidence of fill, other than levees, in assessment area X X
Evidence of fill or channelization upstream of assessment area X
Evidence of fill or channelization downstream of assessment area X
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 467
described relative to the mean FCI scores for each
function (Table 4). Average FCI scores for the two
wetland classes were $ 0.50 for all functions except
for Biogeochemistry for riverine wetlands. An
average FCI score close to 0.50 indicates that the
wetlands are performing, on average, at about 50%
of the level of reference standard condition (a FCI
score of 1.0). Mean scores of 0.84 for the Plant
Community function of riverine wetlands and 0.76
for the Hydrology function of flats indicate that the
wetlands are performing, on average, close to
reference standard condition for those functions.
Second, examination of cumulative distribution
function (CDF) plots provides additional insights
into the condition of the population of wetlands in
both classes. The CDF plots described in this section
are representative of the two classes over the entire
watershed because they incorporate the correction
for sample bias due to site accessibility. The CDF
plots for the Biogeochemistry and Plant Community
functions for flats are close to linear, indicating that
the FCI scores are evenly distributed over the
population (Figure 3b, d). The plots for the
Hydrology and Habitat functions for flats are not
linear; both have distinct rises. In the case of the
Hydrologic function, a sharp rise occurs as FCI
scores approach 1.0, indicating that about 35% of
the population is functioning at the level of reference
standard condition (Figure 3a). Reflecting the sharp
rise in the CDF plot, the median score for hydrology
is 0.92 (Table 4), which is much higher than the
mean of 0.76. This indicates that more than 50% of
the population is functioning at $ 90% of the level
of reference standard. In the case of the Habitat
function, a distinct inflection in the CDF plot occurs
at a FCI score of about 0.43; therefore, about 80%
of the population has a score above 0.43 (Figure 3c).
The CDF plots for riverine wetlands show
a somewhat different pattern for the five functions.
The Hydrology function shows that the assessment
sites fall into two groups, with approximately 60%
above the inflection point at an FCI score of about
0.6, which is also the median score (Table 4). In
contrast, FCI scores are evenly distributed over the
populations for the Biogeochemistry and Landscape
functions (Figure 4b, e), and the median and mean
are similar (Table 4). The relatively steep inflection
points at FCI scores of approximately 0.7 and 0.8
for the Habitat and Plant Community functions
(Figure 4c, d) indicate that the largest proportion of
the population was functioning close to reference
standard.
As described in the Data Analysis section, an
ordination approach was used to compare condi-
tions at the watershed scale further, as well as to
evaluate the relationships among sites and functions
(Figures 5?7). Ordination of FCI scores for the flats
sites were most informative when we divided the
data into two categories (Figures 5 and 6) based on
logging history. The separation of flats into two
groups was based on preliminary data analyses of
the entire data set and our observations that most of
the sites that had been harvested within the past
Table 4. Means and medians of the FCI scores for each function for non-tidal riverine and flats wetlands in the
Nanticoke River watershed. 95% confidence intervals are in parentheses.
Function
Flats Riverine
Mean Median Mean Median
Hydrology 0.76 (0.71, 0.81) 0.92 (0.85, 0.94) 0.53 (0.46, 0.59) 0.60 (0.30, 0.65)
Biogeochemistry 0.54 (0.49, 0.59) 0.59 (0.38, 0.67) 0.44 (0.38, 0.51) 0.42 (0.23, 0.59)
Habitat 0.63 (0.59, 0.67) 0.64 (0.58, 0.70) 0.64 (0.58, 0.71) 0.74 (0.54, 0.83)
Plant Community 0.50 (0.44, 0.56) 0.56 (0.47, 0.63) 0.84 (0.80, 0.88) 0.89 (0.81, 0.91)
Landscape 0.69 (0.65, 0.72) 0.72 (0.68, 0.74)
Figure 3. Cumulative distribution function (CDF) plots
showing the estimated percent of wetland area with FCI
scores less than or equal to any value on the x-axis for
wetlands in the flats class in the Nanticoke River
watershed. CDF plots (6 95% confidence intervals) are
for the FCI scores for the a) Hydrology, b) Biogeochem-
istry, c) Habitat, and d) Plant Community functions
relative to the total area of flats in the watershed.
468 WETLANDS, Volume 27, No. 3, 2007
50 years had also been ditched and the surface soils
were almost always mechanically altered (i.e., heavy
equipment used to clear the site and form windrows
of stumps and soil, etc.). Most of the sites that had
not been altered within the past 50 years, based on
the ages of trees on the site, had been previously
logged but few had been physically altered, although
a few of them had ditches within the assessment
areas or within 1 km of the assessment area.
Another difference between wetlands in the two
categories is that sites that had been altered more
recently are almost always dominated by Loblolly
pine (Pinus taeda L.). While loblolly pine occurs
naturally in non-tidal forested wetlands on the
Eastern Shore (Tiner 1985, Tiner and Burke 1995),
it is most abundant in pine plantations.
The sizes of the circles in the graphs of Figures 5?
7 indicate the relative magnitude of the FCI and DI
scores (e.g., larger circles indicate higher scores and
smaller circles represent lower scores). The ordina-
tion of flats sites that had not been disturbed for the
past 50 years results in two distinct groupings
(Figure 5), and there was no clear pattern in the
magnitude of the DI scores for sites in the two
groups (Figure 5), demonstrating that alterations
that occurred several decades ago continue to
impact the assessment scores for the Hydrology
and Biogeochemistry functions. In contrast, the
Habitat and Plant Community functions differed
little among the sites, most likely because those sites
were dominated by facultative wetland species that
are able to compete successfully under less wet
conditions.
The ordination of sites that had been altered in
the past 50 years resulted in a more complex pattern
(Figure 6), with sites that had been altered in the
past 15 years appearing in the lower portion of the
plot of FCI scores for the Plant Community and, to
a lesser degree, Habitat functions. The plot of DI
values (Figure 6) showed only a general relationship
with the FCI scores with sites to the right of the
ordination corresponding with lower FCI scores for
the Hydrology and Biogeochemistry functions.
The NMS ordination of FCI scores for the five
HGM models for the riverine class resulted in a clear
distribution of sites along the first two axes
(Figure 7). The DI graph in Figure 7 shows that
there was a negative relationship between FCI scores
Figure 4. Cumulative distribution function (CDF) plots
showing the estimated percent of wetland area with FCI
scores less than or equal to any value on the x-axis for
wetlands in the non-tidal riverine class in the Nanticoke
River watershed. CDF plots (6 95% confidence intervals)
are for the FCI scores for the a) Hydrology, b)
Biogeochemistry, c) Habitat, d) Plant Community, and
e) Landscape functions relative to the total area of riverine
wetlands in the watershed.
Figure 5. NMS ordination of FCI scores for the four
HGM functions (Table 2) and the Disturbance Index (DI)
for flats that had no evidence of disturbance during the
past 50 years. In all diagrams, the size of the circle
represents the relative magnitude of the FCI score (larger
circles 5 higher FCI scores or DI values). The two axes
account for 95.1% of the variance and the Hydrology (r 5
0.526), Biogeochemistry (r5 0.366) and Plant Community
(r 5 0.449) are positively and the Habitat function (r 5
-0.633) negatively correlated with Axis 1. The strongest
correlations on Axis 2 are with the Hydrology (r 5 -0.952)
and Biogeochemistry (r 5 -0.965) functions. FCI scores
are provided in Appendix II.
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 469
and the magnitude of the DI. Most sites with high
DI values had low FCI scores.
Condition of Wetlands at the Sub-Basin Scale
There were no significant sub-basin differences in
FCI scores for the flats subclass (Table 5), but there
were significant differences for the riverine sites,
with the Nanticoke River sub-basin having signifi-
cantly lower FCI scores for four of the riverine
functions.
DISCUSSION
Wetlands in both classes had the full range of
variable and function scores (Appendices I and II,
Figures 3?7), indicating that wetland condition at
the watershed scale is less than reference standard
for many assessment sites. The methods that we
have used to analyze the data allow us to determine
how the HGM functions vary across the watershed,
how they vary across the sites sampled, and how
they vary in response to alterations. Our results
clearly confirm predictions of Tiner et al. (2000) and
Tiner (2004, 2005), based on GIS-based analyses of
wetland functions and habitat integrity of the
Nanticoke at the watershed scale, that there likely
has been significant wetland degradation in the
Nanticoke River basin. Tiner (2005) also reached
similar conclusions in a comparison of historical
changes in wetland area and potential losses of
wetland functions in the Nanticoke watershed. Tiner
(2004) also evaluated the Nanticoke watershed from
the perspective of watershed integrity by using
spatial data to evaluate features of the landscape
that would be related to water and wetland quality.
He found that the index of watershed integrity was
low, indicating a ?stressed system? with ?significant
human modification.?
While the objectives of our studies, as well as the
approaches used by our group and by Tiner (2004)
and Tiner et al. (2000), were different, we each came
to similar conclusions, which is not surprising given
the level of human-influenced alterations in the
watershed (TNC 1994, Tiner 2004, 2005). These
Figure 6. NMS ordination of FCI scores for the four
HGM functions (Table 2) and the Disturbance Index (DI)
for flats that had evidence of disturbance during the past
50 years. In all diagrams, the size of the circle represents
the relative magnitude of the FCI score (larger circles 5
higher FCI scores or DI values). The two axes accounted
for 96.0% of the variance, and 76.2% is explained by Axis
1, which is negatively correlated with the Hydrology (r 5
-0.882), Biogeochemistry (r 5 -0.927) and Habitat (r 5
-0.469) functions and positively (r 5 0.616) with the Plant
Community function. The Plant Community and Habitat
functions are positively correlated with Axis 2 (r 5 0.779
and 0.549, respectively). FCI scores are provided in
Appendix II.
Figure 7. NMS ordination of FCI scores for the five
HGM functions (Table 2) and the Disturbance Index (DI)
for non-tidal riverine wetlands. In all diagrams, the size of
the circle represents the relative magnitude of the FCI
score (larger circles 5 higher FCI scores or DI values).
Axis 1 accounted for 95.0% of the variance and it was
highly and positively correlated with the Hydrology (r 5
0.872), Biogeochemistry (r 5 0.901), Habitat (r 5 0.836),
Plant Community (r 5 0.732) and Landscape (r 5 0.564)
functions. Axis 2 accounted for little (4.0%) additional
variance even though all of the functions except Land-
scape (r 5 0.106) were highly correlated with it
(Hydrology: r 5 0.627, Biogeochemistry: r 5 -0.804,
Habitat: r 5 0.845, Plant Community: r 5 -0.610). DI was
positively related to Axis 1 (r 5 0.769) and Axis 2 (r 5
0.755). FCI scores are provided in Appendix I.
470 WETLANDS, Volume 27, No. 3, 2007
alterations include a dense network of channelized
streams and the conversion of a large number of
wetlands into agricultural land uses (Tiner et al.
2000, Tiner 2004, 2005). Wang et al. (1997) also
found that aquatic resources were degraded at the
watershed scale following even relatively small levels
of human-influenced activities (e.g., conversion of
natural areas to agricultural and urban land uses).
Houlahan and Findlay (2004) showed that wetland
water quality also decreases in response to human
activities near wetlands.
In contrast to similar findings at the watershed
scale, results of our study and Tiner?s (2004) study of
watershed integrity differed at the sub-basin scale.
Tiner found that eight of 11 indices of watershed
integrity were lower in the Delaware portion of the
Marshyhope Creek sub-basin (Table 8 in Tiner
2004). In our study, none of the wetland functions
were significantly lower in the Marshyhope sub-
basin for either wetland class, but four of five
functions for riverine wetlands were significantly
lower in the main-stem Nanticoke River sub-basin
(Table 5). The differences between the two studies
could result from differing methodologies. First,
Tiner only evaluated the portion of the sub-basins
that were in Delaware, while we also considered sites
in both states for each sub-basin. Differences in
wetland condition in the two states could lead to
differing assessment results, but the experiences of
our field teams suggest that this factor is not
important. Differences between our results and
Tiner?s also point to results that might be expected
when one works with assessment models for in-
dividual types of wetlands versus models that are not
specific to wetland types. Tiner et al. (2000)
predicted the potential (e.g., high and moderate)
for different types of wetlands to perform 10
functions, but their assessment (Tiner et al. 2000,
Tiner 2004) was not developed to assess function
within individual wetland types (i.e., departure from
??reference??). Our study, based on variables cali-
brated to reference data, indicates that wetland
condition (and we assume wetland functions) is not
the same for flats and riverine wetlands at either the
watershed or sub-basin scales. Third, Tiner?s study
was based on an analysis and interpretation of GIS
data, and there was no attempt to verify his findings
either at the watershed or sub-basin scale and no
consideration of wetland condition (R. Tiner,
personal communication). Tiner recognized the
limitation of the method in making spatial predic-
tions about watershed integrity and suggested that
the models could be improved by comparing
predictions with field-based assessments using field
indicators. Even though the methods differ, our
results support Tiner?s prediction a high level of
wetland alteration in the Nanticoke watershed has
occurred.
The approach that is used to assess wetlands at
the watershed scale will determine how the results
can be used. The landscape approach used in Tiner?s
studies provides an overview of wetland function or
watershed integrity, and it can be used to infer
wetland condition. Assessment of wetland condi-
tion, however, can only be accomplished by
applying field-based assessment models that are
calibrated to a range of wetland condition. This
study and a companion study in the Juniata River
watershed (Wardrop et al. 2007) are the only two of
which we are aware of for which a field-based
approach has been used to assess condition at the
watershed and sub-basin scales. Because of the
statistical method that was used to select our sites
(Stevens and Olsen 1999, 2000), we assume that the
results were representative of the sub-basins and
watershed.
Our study also demonstrates that it is important
to develop models for each type of wetland and to
calibrate the models to reference data. We found
that, to a large degree, the HGM models for the two
wetland classes used different variables (Table 1),
and we did not develop a landscape model for the
Table 5. Results of ANOVA comparisons of mean FCI scores for three Nanticoke subwatersheds. Means that do not
differ significantly share the same superscript at P # 0.03 or better for the riverine sites and P $ 0.05 for the flats sites.
Subwatershed Hydrology Biogeochemistry Habitat Plant Community Landscape
Riverine
Marshyhope Creek (N 5 25) 0.642a 0.543a 0.710a 0.823a 0.762a
Nanticoke River (N 5 11) 0.265b 0.213b 0.401b 0.808a 0.602b
Broad Creek (N 5 12) 0.729a 0.662a 0.869a 0.959b 0.789a
Flats
Marshyhope Creek (N 5 16) 0.715a 0.485a 0.679a 0.529a
Nanticoke River (N 5 33) 0.648a 0.551a 0.668a 0.576a
Broad Creek (N 5 9) 0.756a 0.490a 0.483a 0.574a
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 471
flats subclass because, during the model develop-
ment and testing phases, we could not identify any
landscape metrics that were sensitive enough to
quantify differences in condition between reference
sites along a gradient of alteration. The comparison
also suggests that there may be differences in
condition of wetlands versus condition of the
watershed. In other words, some individual wetlands
may be in good condition despite significant
alteration (e.g., degradation) elsewhere in the
watershed. In a recent study of urbanized portions
of New Jersey, Ehrenfeld (2005) found that relative-
ly few wetlands had plant communities that were
characteristic of reference standard conditions. She
sampled wetlands in five HGM classes (non-tidal
riverine, mineral flats, flats-riverine, depressions,
and slopes) and found a high degree of similarity
in vegetation because hydrologic alterations had
resulted in an increase in the abundance of
facultative wetland plant species, including inva-
sives. As a result, Ehrenfeld was not able to separate
plant communities on the basis of HGM classes,
even though she found that a few sites in the riverine
HGM class had plant communities that were similar
to those at reference standard sites. Ehrenfeld?s
study supports our findings that hydrologic altera-
tions at the watershed scale can have ubiquitous
impacts as demonstrated by low FCI scores for
many of the flats sites. Flats can be affected by
landscape-scale hydrologic alterations such as those
that have occurred in the Nanticoke watershed (i.e.,
rapid removal of surface water and shallow ground
water through a vast network of interconnected
ditches and canals).
The HGM approach that was used in this study
also allows us to compare functions and identify
which ones have been degraded. In riverine wet-
lands, for example, FCI scores were lower for the
Hydrology and Biogeochemistry functions (Ta-
ble 4), indicating that those functions have been
degraded more than the others. Observations by
Tiner (2004) that 79% of the streams were channel-
ized in the Nanticoke watershed support this
interpretation. Higher values for the Habitat, Plant
Community, and Landscape functions for the
riverine subclass (Table 4) also demonstrate that
there can be a wide range of hydrologic changes
before there are noticeable decreases in those
functions. Once established, trees and shrubs in
riverine wetlands may persist despite significant
change in the hydrologic regime by channelization;
the fine-textured soils may be able to retain
sufficient moisture to support these species and
thereby retain Habitat functions. In contrast, the
FCI scores for the Hydrology function in the flats
subclass (Table 4) were higher for many sites
compared to the FCI scores for the other functions.
These results suggest that the vegetation responds
more dramatically to the variety of stresses (e.g.,
physical disturbance associated with logging activ-
ities, physical differences in soil properties resulting
in variation in moisture retention capacity). There
are also other possible interpretations for the
relatively high FCI scores for the Hydrology
function in the flats subclass. The model used in
our study was adapted from the hydrology model
for wet pine flats developed by Rheinhardt et al.
(2002) for an area bounded by the Neuse River
(North Carolina) to coastal northeast Florida and
the Big Thicket area of southeastern Texas. The
model developed by Rheinhardt et al. also was
adapted for use in hardwood flats in the Coastal
Plain of Virginia (Havens et al. 2001). We were
unable to make direct hydrologic measurements to
verify the applicability of the variables that Rhein-
hardt et al. used in their models for our more
northern sites, but given the lower FCI scores for the
Plant Community and Habitat functions, it seems
likely that the model overestimated the Hydrology
function of flats in the Nanticoke watershed.
Further evaluation of the Hydrology function
should be conducted with emphasis on the hydro-
logic alterations that are caused by ditches of
varying sizes. Another possible explanation for our
results is that extensive ditching within the water-
shed also has affected hydrologic conditions in
reference standard sites against which all sites were
calibrated. If this situation exists, calibration of the
Hydrology model of Rheinhardt et al. with further
data on the influences that ditches of varying sizes
have on soil moisture and ground-water levels would
be required to accurately compare assessment sites
and reference standard sites.
In summary, we found that the HGM method can
be applied to an entire watershed to evaluate
wetland condition when it is used in combination
with the EMAP approach for site selection. While
time-consuming (e.g., the amount of time to develop
the models, obtain access to sites, conduct the
assessments), this approach has several benefits.
First, it provides HGM models that are based on
reference data, an important aspect of any assess-
ment model (Brinson et al. 1995, Smith et al. 1995).
The HGM models are available and can be used to
assess individual sites as part of wetland assessments
for permit applications. Second, the HGM models
should be applicable in other parts of the Outer
Coastal Plain where there are similar topographic
conditions and similar vegetation. In this context,
the flats models developed by Havens et al. (2001)
472 WETLANDS, Volume 27, No. 3, 2007
and Rheinhardt et al. (2002) are extended further
north on the Atlantic Coastal Plain. Third, the
approach that we have used in this study demon-
strates that it is possible to identify portions of
a watershed (i.e., subwatersheds) where wetland
condition is either close to reference standard
condition or where condition indicates that signifi-
cant degradation has occurred. In instances where
wetlands are close to reference standard condition,
managers should focus on conservation. In areas
where there is evidence of significant degradation,
managers should focus on restoration. The results of
this study can also be used to assist managers in
restoration activities because the reference data can
provide the basis for developing guidelines and
criteria for restoration of specific sites, as well as
data that can be used to track restoration success.
ACKNOWLEDGMENTS
The study was funded under cooperative agree-
ment #82681701 to the Smithsonian Environmental
Research Center from the EPA through the
National Health and Environmental Effects Labo-
ratory ? Western Ecology Division in Corvallis,
Oregon. The information in this document has been
subjected to review by EPA and approved for
publication. Approval does not signify that the
contents reflect the views of the Agency, nor does
mention of trade names or commercial products
constitute endorsement or recommendation for use.
The efforts of Richard Sumner and Art Spingarn
(deceased), EPA cooperators on the project, are
appreciated, especially their support in making the
project happen and in generating interest in the
results. The individuals who formed the field teams
are greatly appreciated. We especially thank Chris
Bason, Stephanie Blades, Pat Groller, Jeff Lin, Bill
Reybold, Christine Whitcraft, and Mike Yarcusko.
Individuals who participated in the workshops that
were used to develop, calibrate, and test the HGM
models were David Bleil, Mark Brinson, Robert
Brooks Leander Brown, Kirk Havens, Rick Re-
inhardt, Charles Rhodes, Art Spingarn, and Denice
Heller Wardrop. We thank two anonymous re-
viewers, as well as Bill Ainslie, Mark Brinson, Rick
Rheinhardt, and especially Ralph Tiner for improv-
ing the manuscript by providing thoughtful com-
ments and suggestions.
LITERATURE CITED
Abbruzzese, B. and S. Leibowitz. 1997. A synoptic approach for
assessing cumulative impacts to wetlands. Environmental
Management 21:457?75.
Bartoldus, C. C. 1999. A Comprehensive Review of Wetland
Assessment Procedures: A Guide for Wetland Practitioners.
Environmental Concern, St. Michaels, MD, USA.
Bedford, B. 1996. The need to define hydrologic equivalence at
the landscape scale for freshwater wetland mitigation. Ecolog-
ical Applications 6:241?51.
Brinson, M. M., F. R. Hauer, L. C. Lee, W. L. Nutter, R. D.
Rheinhardt, R. D. Smith, and D. Whigham. 1995. A
Guidebook for Application of Hydrogeomorphic Assessments
to Riverine Wetlands. U.S. Army Corps of Engineers, Water-
ways Experiment Station, Vicksburg, MS, USA.Technical
Report WRP-DE-11.
Brooks, R. P., D. Heller Wardrop, and J. A. Bishop. 2004.
Assessing wetland condition on a watershed basis in the Mid-
Atlantic region using synoptic land cover maps. Environmental
Monitoring and Assessment 94:9?22.
Cormier, S. M., M. Smith, S. Norton, and T. Neiheisel. 2000.
Assessing ecological risk in watersheds: a case study of problem
formulation in the Big Darby Creek watershed, Ohio, USA.
Environmental Toxicology and Chemistry 19:1082?96.
Dalenius, T., J. Ha?jek, and S. Zubrzycki. 1961. On plane
sampling and related geometrical problems. Proceedings of
the 4th Berkeley Symposium on Probability and Mathematical
Statistics 1:125?50.
Detenbeck, N. E., S. L. Batterman, V. J. Brady, J. C. Brazner, V.
M. Snarski, D. L. Taylor, J. A. Thompson, and J. W. Arthur.
2000. A test of watershed classification systems for ecological
risk assessment. Environment, Toxicology and Chemistry
19:1174?81.
Ehrenfeld, J. G. 2005. Vegetation of forested wetlands in turban
and suburban landscapes in New Jersey. Journal of the Torrey
Botanical Society 132:262?79.
Fennessy, M. S., A. D. Jacobs, and M. E. Kentula. 2004. Review
of Rapid Methods for Assessing Wetland Condition. U.S.
Environmental Protection Agency, Washington, DC,
USA.EPA/620/R-04/009.
Havens, K. J., C. Coppock, R. Arenson, D. Stanhope, and G.
Silberhorn. 2001. A Draft Regional Guidebook for Applying
the Hydrogeomorphic Approach to Wet Hardwood Flats on
Mineral Soils in the Coastal Plain of Virginia. Draft document
prepared for the U.S. Environmental Protection Agency,
Region III, Virginia Institute of Marine Sciences, Glouster,
VA, USA.
Houlahan, J. E. and C. S. Findlay. 2004. Estimating the ?critical?
distance at which adjacent land-use degrades wetland water and
sediment quality. Landscape Ecology 19:677?90.
Jordan, T. E., M. A. Pittek, R. P. Szuch, D. F. Whigham, and D.
E. Weller. 2007. Comparing rapid functional assessments of
wetlands to measurements of soil characteristics and nitrogen
processing. Wetlands 27:479?97.
Lee, L. C., P. L. Fiedler, S. R. Steward, D. J. Partridge, J. A.
Mason, E. M. Inlander, and M. C. Rains. 2003. Draft
Operational Guidebook for Assessment of the Functions of
Riverine Waters/Wetlands in the Santa Margarita Watershed,
Riverside & San Diego Counties, California. San Diego
Regional Water Quality Control Board, Technical Publication,
San Diego, CA, USA.
Lesser, V. M. 2001. Applying survey research methods to account
for denied access to research sites on private property.
Wetlands 21:639?47.
Leibowitz, S. G., C. Loehle, B-L. Li, and E. M. Preston. 2000.
Modeling landscape functions and effects: a network approach.
Ecological Modelling 132:77?94.
Lemly, A. D. 1997. Risk assessment as an environmental
management tool: considerations for freshwater wetlands.
Environmental Management 21:85?102.
McCune, B. and M. J. Mefford. 1999. PC-ORD. Multivariate
Analysis of Ecological Data, Version 4. MJM Software Design,
Gleneden Beach, OR, USA.
Montgomery, D. R., G. E. Grant, and K. Sullivan. 1995.
Watershed analysis as a framework for implementing ecosys-
tem management. Water Resources Bulletin 31:369?86.
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 473
Overton, W. S. and S. V. Stehman. 1993. Properties of designs for
sampling continuous spatial resources from a Triangular Grid.
Communications in Statistics Part A: Theory and Methods
22:2641?60.
Radford, A. E., H. E. Ahles, and C. R. Bell. 1968. Manual of the
Vascular Flora of the Carolinas. The University of North
Carolina Press, Chapel Hill, NC, USA.
Rheinhardt, R. D., M. C. Rheinhardt, and M. M. Brinson. 2002.
A Regional Guidebook for Applying the Hydrogeomorphic
Approach to Assessing Wetland Functions of Wet Pine Flats
on Mineral Soils in the Atlantic and Gulf Coastal Plains. U.S.
Army Engineer Research and Development Center, Vicksburg,
MS, USA.Technical Report ERDC/ELTR-02-9.
SAS Institute, Inc. 2004. SAS/STAT 9.1 User?s Guide. SAS
Institute, Inc., Cary, NC, USA.
Smith, R. D., A. Ammann, C. Bartoldus, and M. M. Brinson.
1995. An Approach for Assessing Wetland Functions Using
Hydrogeomorphic Classification, Reference Wetlands, and
Functional Indices. U.S. Army Corps of Engineers, Waterways
Experiment Station, Vicksburg, MS, USA.Wetland Research
Program Technical Report WRP-DE.
State of Delaware. 1994. Procedures Manual: Development of
Statewide Digital Orthophotography, Street Centerlines, and
Wetlands Mapping. Prepared for the Delaware Dept. of
Transportation and DNREC by PhotoScience, Inc., Dewberry
& Davis, and Environmental Research, Inc. PhotoScience, Inc.,
Gaithersburg, MD, USA.
Stevens, D. L., Jr. 1997. Variable density grid-based sampling
designs for continuous spatial populations. Environmetrics
8:167?95.
Stevens, D. L., Jr. and S. F. Jensen. 2007. Sampling design,
implementation, and analysis for wetland assessment. Wetlands
27:515?23.
Stevens, D. L., Jr. and A. R. Olsen. 1999. Spatially restricted
surveys over time for aquatic resources. Journal of Agricultur-
al, Biological, and Environmental Statistics 4:415?28.
Stevens, D. L., Jr. and A. R. Olsen. 2000. Spatially-restricted
random sampling designs for design-based and model-based
estimation. p. 609?16. In Accuracy 2000: Proceedings of the 4th
International Symposium on Spatial Accuracy Assessment in
Natural Resources and Environmental Sciences. Delft Univer-
sity Press, Delft, The Netherlands.
Sutter, L., J. Sanfill, D. Haupt, C. Bruce, and J. Wuenscher. 1999.
NC-CREWS: North Carolina Coastal Region Evaluation of
Wetland Significance, a report of the Strategic Plan for
Improving Coastal Management in North Carolina. North
Carolina Department of Environment and Natural Resources,
Raleigh, NC, USA.
The Nature Conservancy. 1994. Nanticoke-Blackwater River
Bioreserve Strategic Plan. Draft 6-10-94. Maryland Nature
Conservancy, Chevy Chase, MD, USA.
Thomas, R. and Z. Lamb. 2004. Scientific perspectives on
a watershed approach to compensatory mitigation. National
Wetlands Newsletter. September-October: 17?20.
Tiner, R. W. 1985. Wetlands of Delaware. U.S. Fish and Wildlife
Service, National Wetlands Inventory, Newton Corner, MA
and Delaware Department of Natural Resources and Environ-
mental Control, Wetlands Section, Dover, DE, USA.
Tiner, R. W. 2004. Remotely-sensed indicators for monitoring the
general condition of ?natural habitat? in watersheds: an
application for Delaware?s Nanticoke River watershed. Eco-
logical Indicators 4:227?43.
Tiner, R. W. 2005. Assessing cumulative loss of wetland function
in the Nanticoke river watershed using enhanced national
Wetlands Inventory data. Wetlands 25:405?19.
Tiner, R. W. and D. G. Burke. 1995. Wetlands of Maryland. U.S.
Fish and Wildlife Service, National Wetlands Inventory,
Newton Corner, MA and Maryland Department of Natural
Resources, Annapolis, MD, USA.
Tiner, R., M. Starr, H. Bergquist, and J. Swords. 2000.
Watershed-based Wetland Characterization for Maryland?s
Nanticoke River and Coastal Bays Watersheds: A Preliminary
Assessment Report. U.S. Fish & Wildlife Service, National
Wetlands Inventory (NWI) Program, Northeast Region,
Hadley, MA, USA.
Wang, L., J. Lyons, P. Kanehl, and R. Gatti. 1997. Influences of
watershed land use on habitat quality and biotic integrity of
Wisconsin stream. Fisheries 22:6?12.
Wardrop, D. H., M. E. Kentula, D. L. Stevens, Jr., S. F. Jensen,
and R. P. Brooks. 2007. Assessment of wetland condition: an
example from the Upper Juniata Watershed in Pennsylvania,
USA. Wetlands 27:416?31.
Weller, D. E., M. N. Snyder, D. F. Whigham, A. D. Jacobs, and
T. E. Jordan. 2007. Landscape indicators of wetland condition
in the Nanticoke River watershed. Wetlands 27:498?514.
Whigham, D. F., D. E. Weller, A. Deller Jacobs, T. E. Jordan,
and M. E. Kentula. 2003. Assessing the ecological condition of
wetlands at the catchment scale. Landschap 2:99?112.
Whigham, D. F., D. E. Weller, and T. E. Jordan. 2000. Quality
Assurance Project Plan for Nanticoke Wetland Assessment
Study. U.S. Environmental Protection Agency, Internal Re-
port, June 2000, Corvallis, OR, USA.
Whittier, T. R., S. G. Paulsen, D. P. Larsen, S. A. Peterson, A. T.
Herilhy, and P. R. Kaufmann. 2002. Indicators of ecological
stress and their extent in the population of Northeastern lakes:
a regional-scale assessment. Bioscience 52:235?47.
Zampella, R., R. Lathrop, J. A. Bognar, L. Craig, and K. Landig.
1994. A watershed-based Wetland Assessment Method for the
New Jersey Pinelands, NJ Pinelands Commission. New Lisbon,
NJ, USA.
Manuscript received 21 November 2005; accepted 24 January
2007.
474 WETLANDS, Volume 27, No. 3, 2007
Appendix I. Variable and FCI scores for sites in the Riverine subclass.
S
IT
E
V
F
A
R
B
U
F
F
E
R
V
F
L
O
O
D
P
L
A
IN
V
IN
V
A
S
IV
E
V
N
E
A
R
B
U
F
F
E
R
V
S
A
P
L
IN
G
V
S
H
R
U
B
V
S
T
R
E
A
M
IN
V
S
T
R
E
A
M
O
U
T
V
T
B
A
V
T
D
E
N
V
T
R
E
E
V
D
IS
T
U
R
B
H
yd
ro
lo
gy
B
io
ge
o
ch
em
is
tr
y
H
ab
it
at
P
la
n
t
C
o
m
m
u
n
it
y
L
an
d
sc
ap
e
22 0.50 0.75 0.75 0.97 1.00 0.10 1.00 0.50 1.00 0.70 1.00 1.00 0.65 0.65 0.83 0.94 0.74
30 0.56 1.00 1.00 1.00 1.00 1.00 1.00 0.75 1.00 0.84 1.00 1.00 0.87 0.87 0.99 1.00 0.83
34 0.89 0.25 1.00 1.00 1.00 1.00 1.00 1.00 0.45 0.84 1.00 1.00 0.71 0.32 0.94 1.00 0.97
36 0.56 0.75 0.94 0.93 0.75 1.00 1.00 1.00 0.85 0.60 1.00 1.00 0.91 0.78 0.95 0.89 0.86
73 1.00 0.25 1.00 0.99 0.90 1.00 1.00 1.00 1.00 0.91 0.90 1.00 0.71 0.71 0.99 0.93 1.00
123 0.56 1.00 1.00 0.92 0.75 1.00 0.10 0.10 0.81 1.00 0.75 1.00 0.26 0.21 0.53 0.81 0.63
135 0.13 0.75 1.00 1.00 0.75 1.00 0.75 0.75 0.80 0.74 0.75 0.50 0.75 0.60 0.75 0.81 0.72
165 0.88 0.10 1.00 0.98 1.00 0.10 0.10 0.10 1.00 1.00 1.00 0.75 0.10 0.10 0.36 1.00 0.74
186 0.31 0.25 1.00 0.54 0.90 1.00 1.00 0.50 0.79 1.00 0.90 1.00 0.50 0.40 0.98 0.93 0.47
197 0.52 0.75 1.00 1.00 0.75 1.00 0.10 0.10 0.79 0.84 0.50 1.00 0.23 0.18 0.52 0.72 0.66
211 1.00 1.00 1.00 1.00 0.90 1.00 1.00 0.75 1.00 1.00 0.75 1.00 0.87 0.87 1.00 0.87 0.94
237 0.94 0.75 1.00 0.95 1.00 1.00 0.75 0.50 0.84 1.00 0.90 1.00 0.61 0.51 0.86 0.96 0.84
241 1.00 0.75 0.92 1.00 0.75 0.10 0.10 0.10 1.00 0.84 0.90 1.00 0.23 0.23 0.39 0.85 0.78
261 0.56 0.75 1.00 0.78 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.91 0.91 1.00 1.00 0.78
264 0.53 0.75 1.00 0.93 0.75 1.00 0.10 0.10 1.00 0.91 0.90 1.00 0.23 0.23 0.54 0.87 0.62
277 0.81 0.25 0.89 0.91 0.75 1.00 0.50 0.50 0.91 1.00 0.75 1.00 0.41 0.37 0.74 0.79 0.78
301 0.13 1.00 1.00 0.81 0.75 1.00 1.00 1.00 0.72 0.63 0.90 1.00 1.00 0.72 0.95 0.87 0.69
309 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.88 0.98 1.00 1.00 1.00 0.88 0.99 1.00 1.00
334 0.56 1.00 1.00 1.00 1.00 0.10 1.00 1.00 1.00 0.67 1.00 1.00 1.00 1.00 0.82 1.00 0.89
355 0.48 0.75 1.00 0.97 1.00 0.61 1.00 0.50 1.00 1.00 1.00 1.00 0.65 0.65 0.94 1.00 0.73
391 0.50 0.25 0.66 0.50 0.25 0.44 0.10 0.10 0.16 0.28 0.90 0.10 0.14 0.02 0.18 0.60 0.40
409 0.56 1.00 1.00 0.92 1.00 1.00 1.00 0.50 1.00 1.00 1.00 1.00 0.71 0.71 1.00 1.00 0.73
416 0.67 0.25 0.98 1.00 1.00 1.00 0.25 0.10 0.76 0.74 1.00 0.25 0.16 0.12 0.46 1.00 0.69
423 0.50 0.25 0.94 0.87 0.50 0.36 0.10 0.10 0.90 0.63 0.90 0.25 0.14 0.13 0.28 0.76 0.59
436 1.00 1.00 1.00 1.00 0.75 1.00 1.00 1.00 1.00 0.95 1.00 1.00 1.00 1.00 1.00 0.91 1.00
440 0.88 0.75 1.00 0.86 1.00 0.86 1.00 0.75 1.00 0.98 1.00 1.00 0.79 0.79 0.98 1.00 0.84
506 0.69 0.75 1.00 0.69 0.75 1.00 0.75 0.75 1.00 1.00 1.00 0.50 0.75 0.75 0.79 0.91 0.71
551 0.56 1.00 1.00 0.96 0.75 0.34 1.00 1.00 1.00 0.77 0.75 1.00 1.00 1.00 0.87 0.81 0.87
570 0.13 0.25 0.88 0.42 0.50 0.38 0.10 0.10 0.95 0.39 0.75 0.25 0.14 0.13 0.27 0.69 0.27
572 0.44 1.00 1.00 0.85 1.00 0.36 0.50 1.00 0.47 0.35 1.00 1.00 0.91 0.43 0.55 1.00 0.79
576 1.00 1.00 0.95 1.00 0.75 0.10 1.00 1.00 0.82 0.70 0.75 1.00 1.00 0.82 0.81 0.80 1.00
595 0.56 0.25 0.92 1.00 0.25 0.15 0.10 0.10 0.68 0.70 0.50 0.25 0.14 0.10 0.23 0.51 0.67
596 0.56 0.75 1.00 0.85 1.00 0.50 0.75 0.50 0.81 0.42 1.00 1.00 0.61 0.50 0.73 1.00 0.69
646 0.44 0.75 1.00 0.73 0.75 0.40 0.10 0.10 0.49 0.56 1.00 1.00 0.23 0.11 0.37 0.91 0.50
697 0.94 0.75 1.00 0.87 0.90 0.69 1.00 1.00 0.59 0.84 1.00 1.00 0.91 0.54 0.90 0.96 0.92
734 0.50 1.00 1.00 1.00 0.75 1.00 1.00 1.00 1.00 0.67 1.00 1.00 1.00 1.00 0.97 0.91 0.88
739 0.94 0.10 1.00 0.94 0.75 0.10 0.10 0.10 0.10 0.10 0.75 0.10 0.10 0.01 0.10 0.81 0.73
747 1.00 0.75 1.00 1.00 0.50 1.00 0.75 0.50 0.82 0.91 0.75 0.75 0.61 0.50 0.81 0.72 0.88
752 0.13 0.25 0.91 0.66 0.25 0.10 0.10 0.10 0.69 0.53 0.25 0.25 0.14 0.10 0.21 0.42 0.39
784 0.92 1.00 1.00 0.88 1.00 1.00 1.00 0.50 1.00 1.00 1.00 1.00 0.71 0.71 1.00 1.00 0.80
786 0.58 1.00 1.00 0.97 1.00 0.21 1.00 1.00 1.00 0.84 1.00 1.00 1.00 1.00 0.86 1.00 0.88
797 1.00 0.25 0.83 0.95 0.25 0.10 0.10 0.10 0.61 0.53 0.75 0.10 0.14 0.09 0.18 0.58 0.75
811 1.00 1.00 1.00 1.00 0.90 0.10 1.00 1.00 1.00 0.63 1.00 1.00 1.00 1.00 0.82 0.96 1.00
821 0.75 0.25 0.87 0.75 0.50 0.13 0.10 0.10 0.96 0.63 0.75 0.25 0.14 0.14 0.25 0.69 0.59
823 0.19 0.75 1.00 0.60 1.00 1.00 1.00 0.50 0.39 0.70 1.00 1.00 0.65 0.25 0.92 1.00 0.47
827 1.00 0.75 0.95 0.95 0.75 0.10 1.00 0.50 1.00 1.00 0.75 1.00 0.65 0.65 0.85 0.80 0.85
830 0.53 0.25 0.97 0.84 0.75 0.31 1.00 0.75 0.94 0.60 1.00 0.25 0.61 0.58 0.72 0.90 0.74
838 0.34 0.25 1.00 0.47 0.75 0.61 1.00 0.50 1.00 0.70 1.00 1.00 0.50 0.50 0.91 0.91 0.45
848 0.53 1.00 1.00 0.83 0.75 0.36 0.10 0.10 0.88 0.56 0.75 0.50 0.26 0.23 0.31 0.81 0.57
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 475
S
IT
E
V
F
A
R
B
U
F
F
E
R
V
F
L
O
O
D
P
L
A
IN
V
IN
V
A
S
IV
E
V
N
E
A
R
B
U
F
F
E
R
V
S
A
P
L
IN
G
V
S
H
R
U
B
V
S
T
R
E
A
M
IN
V
S
T
R
E
A
M
O
U
T
V
T
B
A
V
T
D
E
N
V
T
R
E
E
V
D
IS
T
U
R
B
H
yd
ro
lo
gy
B
io
ge
o
ch
em
is
tr
y
H
ab
it
at
P
la
n
t
C
o
m
m
u
n
it
y
L
an
d
sc
ap
e
889 0.25 0.25 0.50 0.52 0.25 0.10 0.10 0.10 0.10 0.10 0.50 0.10 0.14 0.01 0.10 0.41 0.35
905 0.94 1.00 1.00 0.76 0.75 0.82 1.00 0.50 0.42 0.70 1.00 0.75 0.71 0.30 0.86 0.91 0.74
930 0.38 1.00 1.00 0.71 0.90 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.96 0.70
934 0.78 1.00 1.00 1.00 1.00 0.10 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.85 1.00 0.95
1045 0.13 1.00 1.00 0.89 0.75 0.10 1.00 0.50 0.88 0.84 0.75 1.00 0.71 0.62 0.83 0.81 0.60
Appendix I. Continued.
476 WETLANDS, Volume 27, No. 3, 2007
Appendix II. Variable Scores and Functional Index Scores (FCI) for assessed sites in the Flats wetland class. Sites that
were disturbed within the past 50 years are indicated in bold, and sites that were not disturbed within the past 50 years are
indicated in italics. Variable scores appear in columns 2?12, and FCI scores in the last four columns. Descriptions of
variables and equations used to calculate FCI scores are given in Tables 1 and 2, respectively.
S
IT
E
V
D
IS
T
U
R
B
V
D
R
A
IN
V
F
IL
L
V
H
E
R
B
V
M
IC
R
O
V
T
R
E
E
V
R
U
B
U
S
V
S
H
R
U
B
V
S
N
A
G
V
T
B
A
V
T
D
E
N
H
yd
ro
lo
gy
B
io
ge
o
ch
em
is
tr
y
H
ab
it
at
P
la
n
t
C
o
m
m
u
n
it
y
6 0.50 0.42 0.50 0.25 0.50 0.50 0.75 0.23 0.10 0.30 0.66 0.44 0.19 0.33 0.28
7 1.00 0.52 1.00 0.75 1.00 1.00 1.00 0.10 1.00 1.00 1.00 0.64 0.64 0.78 0.88
12 1.00 1.00 1.00 0.75 1.00 0.25 1.00 1.00 0.10 1.00 1.00 1.00 0.85 0.78 0.50
20 1.00 0.10 1.00 0.75 1.00 0.25 1.00 1.00 1.00 1.00 1.00 0.33 0.33 1.00 0.50
24 0.75 1.00 1.00 0.75 0.75 0.75 0.75 1.00 1.00 1.00 0.81 1.00 0.84 0.91 0.56
33 0.75 1.00 1.00 0.75 0.75 1.00 1.00 0.10 0.50 0.94 1.00 1.00 0.78 0.58 0.88
50 0.50 1.00 0.75 0.75 0.50 1.00 0.75 1.00 0.10 1.00 1.00 0.94 0.56 0.65 0.66
55 0.75 1.00 1.00 0.75 0.75 1.00 1.00 1.00 1.00 0.90 1.00 1.00 0.86 0.93 0.88
60 1.00 0.10 0.75 0.75 1.00 1.00 1.00 0.34 1.00 1.00 1.00 0.26 0.26 0.84 0.88
64 0.50 0.10 0.75 0.10 0.25 0.25 0.10 0.10 1.00 0.60 0.76 0.26 0.14 0.57 0.02
66 1.00 0.16 1.00 0.75 1.00 0.50 1.00 1.00 1.00 1.00 1.00 0.37 0.37 1.00 0.63
68 0.75 1.00 1.00 0.75 0.75 1.00 1.00 1.00 1.00 0.55 0.95 1.00 0.79 0.88 0.88
70 0.50 1.00 0.50 0.75 0.75 0.25 1.00 1.00 1.00 1.00 0.80 0.88 0.74 0.85 0.50
76 0.50 0.55 0.75 0.25 0.10 0.25 0.75 1.00 1.00 0.27 0.76 0.60 0.23 0.75 0.19
80 1.00 0.95 1.00 0.75 1.00 0.50 1.00 1.00 1.00 1.00 1.00 0.96 0.96 1.00 0.63
84 0.50 0.71 1.00 0.25 0.75 0.25 0.50 0.13 0.10 0.10 0.10 0.78 0.33 0.21 0.13
91 0.75 0.99 1.00 0.75 0.50 0.25 1.00 1.00 0.10 1.00 1.00 0.99 0.60 0.71 0.50
93 0.75 0.10 0.75 0.25 0.10 0.50 0.10 0.10 0.72 0.94 1.00 0.26 0.13 0.64 0.04
99 0.25 0.10 0.50 0.50 1.00 1.00 0.75 0.50 1.00 1.00 1.00 0.20 0.20 0.69 0.56
129 0.50 1.00 1.00 0.25 0.50 0.25 0.10 1.00 0.50 0.10 0.10 1.00 0.37 0.53 0.03
138 0.75 1.00 0.75 0.25 0.75 1.00 1.00 1.00 0.10 0.47 1.00 0.94 0.60 0.65 0.63
140 0.75 0.72 1.00 0.75 1.00 0.75 1.00 0.10 0.10 0.37 0.76 0.79 0.56 0.38 0.75
141 1.00 0.10 1.00 1.00 1.00 0.50 0.75 1.00 1.00 1.00 1.00 0.33 0.33 1.00 0.56
142 1.00 1.00 1.00 0.75 0.75 1.00 0.75 0.94 0.50 1.00 1.00 1.00 0.79 0.86 0.66
144 0.75 0.79 0.75 0.75 0.75 1.00 0.75 1.00 0.10 0.58 0.62 0.78 0.46 0.61 0.66
147 0.50 1.00 1.00 0.50 0.50 0.10 0.10 0.10 0.10 0.10 0.10 1.00 0.30 0.20 0.03
149 1.00 0.80 1.00 1.00 1.00 1.00 1.00 1.00 0.10 0.79 0.81 0.85 0.67 0.73 1.00
172 1.00 1.00 0.75 0.75 0.75 1.00 0.75 0.57 0.50 1.00 0.86 0.94 0.72 0.75 0.66
196 0.75 0.80 1.00 0.75 0.75 0.75 0.75 0.36 0.50 1.00 1.00 0.85 0.67 0.65 0.56
226 0.75 0.10 0.75 0.50 0.75 0.50 1.00 1.00 0.10 0.17 0.33 0.26 0.12 0.53 0.50
229 0.75 0.82 1.00 0.75 0.75 1.00 1.00 0.80 0.10 1.00 1.00 0.87 0.63 0.66 0.88
230 0.50 0.50 0.75 0.25 0.50 0.25 0.50 1.00 0.10 0.10 0.10 0.56 0.17 0.43 0.13
242 0.50 1.00 1.00 0.50 0.75 0.25 0.50 0.87 1.00 0.42 1.00 1.00 0.78 0.77 0.19
245 0.75 1.00 0.75 0.75 0.75 0.75 1.00 1.00 1.00 0.85 1.00 0.94 0.80 0.92 0.75
246 0.75 1.00 1.00 0.75 0.10 0.50 0.75 0.54 0.10 1.00 1.00 1.00 0.40 0.60 0.47
249 0.25 0.21 0.50 0.75 0.75 0.50 1.00 1.00 0.10 1.00 1.00 0.28 0.20 0.59 0.63
274 1.00 0.79 1.00 0.75 1.00 0.50 0.10 1.00 1.00 1.00 1.00 0.84 0.84 1.00 0.06
297 0.75 0.63 0.50 0.75 0.75 0.50 0.25 0.10 0.50 0.46 0.61 0.60 0.38 0.47 0.16
303 0.75 0.93 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.81 1.00 0.95 0.92 0.91 1.00
304 0.75 1.00 0.50 0.75 0.10 0.50 0.10 1.00 0.10 0.83 1.00 0.88 0.33 0.69 0.06
307 0.50 1.00 0.75 0.75 0.75 0.25 0.10 0.54 0.10 0.49 1.00 0.94 0.60 0.47 0.05
317 0.50 0.10 0.75 0.50 0.25 0.10 0.10 0.40 0.50 0.10 0.10 0.26 0.06 0.38 0.03
318 0.75 0.85 0.75 0.75 1.00 1.00 1.00 0.17 1.00 1.00 1.00 0.83 0.83 0.73 0.88
329 0.75 0.10 0.75 0.25 0.10 0.25 0.50 0.17 0.10 0.92 1.00 0.26 0.10 0.50 0.13
330 0.75 1.00 1.00 0.75 1.00 0.50 1.00 1.00 1.00 0.72 0.71 1.00 0.91 0.87 0.63
335 0.75 0.10 0.50 0.75 0.50 1.00 0.10 0.23 0.10 1.00 0.95 0.20 0.12 0.51 0.09
Whigham et al., HGM WETLAND ASSESSMENT AT THE WATERSHED SCALE 477
S
IT
E
V
D
IS
T
U
R
B
V
D
R
A
IN
V
F
IL
L
V
H
E
R
B
V
M
IC
R
O
V
T
R
E
E
V
R
U
B
U
S
V
S
H
R
U
B
V
S
N
A
G
V
T
B
A
V
T
D
E
N
H
yd
ro
lo
gy
B
io
ge
o
ch
em
is
tr
y
H
ab
it
at
P
la
n
t
C
o
m
m
u
n
it
y
340 0.75 1.00 0.50 0.75 0.75 1.00 1.00 0.10 0.50 0.77 1.00 0.88 0.66 0.56 0.88
342 0.75 1.00 1.00 0.25 0.10 0.25 0.50 0.10 0.10 0.46 1.00 1.00 0.31 0.42 0.13
345 0.75 1.00 1.00 0.50 0.75 0.50 1.00 1.00 1.00 0.57 0.67 1.00 0.75 0.84 0.50
354 1.00 0.21 1.00 0.75 1.00 0.50 1.00 1.00 0.10 0.11 0.33 0.41 0.24 0.58 0.63
367 0.10 1.00 1.00 0.50 1.00 0.10 0.75 0.10 0.50 0.10 0.10 1.00 0.62 0.20 0.23
368 1.00 1.00 1.00 0.75 0.75 1.00 1.00 1.00 1.00 0.96 1.00 1.00 0.87 1.00 0.88
369 1.00 1.00 1.00 0.75 1.00 0.75 1.00 1.00 0.10 0.88 1.00 1.00 0.83 0.76 0.75
374 1.00 1.00 0.75 0.75 1.00 0.25 1.00 1.00 1.00 0.87 0.81 0.94 0.89 0.96 0.50
379 0.75 0.16 0.75 0.75 0.10 0.25 0.10 1.00 0.10 1.00 1.00 0.31 0.12 0.71 0.05
387 1.00 0.10 1.00 0.75 1.00 1.00 1.00 1.00 0.50 1.00 1.00 0.33 0.30 0.88 0.88
393 1.00 0.11 1.00 0.75 1.00 1.00 1.00 0.30 0.10 1.00 1.00 0.33 0.28 0.60 0.88
396 1.00 1.00 1.00 0.75 1.00 1.00 1.00 1.00 1.00 1.00 0.86 1.00 0.98 0.98 0.88
399 1.00 1.00 0.75 1.00 1.00 1.00 0.75 0.27 0.50 1.00 1.00 0.94 0.86 0.69 0.75
400 0.75 1.00 1.00 0.75 1.00 1.00 1.00 0.10 0.10 1.00 1.00 1.00 0.85 0.49 0.88
401 0.75 0.32 0.75 0.75 0.75 1.00 1.00 1.00 1.00 1.00 0.95 0.43 0.37 0.93 0.88
414 0.25 1.00 0.75 1.00 0.75 1.00 1.00 0.10 0.50 0.96 1.00 0.94 0.74 0.46 1.00
417 0.75 0.89 1.00 0.25 0.10 0.25 0.10 0.30 0.10 0.36 0.95 0.92 0.26 0.45 0.03
424 0.75 0.73 0.75 0.50 1.00 0.75 0.75 0.10 0.50 1.00 1.00 0.74 0.67 0.59 0.47
430 0.10 1.00 1.00 1.00 0.75 1.00 1.00 1.00 1.00 0.66 1.00 1.00 0.82 0.73 1.00
431 0.75 1.00 0.75 0.75 0.75 1.00 1.00 0.60 1.00 1.00 1.00 0.94 0.82 0.84 0.88
434 0.10 0.82 1.00 0.25 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.87 0.09 0.10 0.02
437 0.75 1.00 1.00 0.75 1.00 1.00 1.00 0.10 0.10 0.99 1.00 1.00 0.85 0.49 0.88
441 1.00 1.00 1.00 1.00 1.00 0.75 1.00 1.00 0.50 1.00 1.00 1.00 0.92 0.88 0.88
442 0.50 0.52 0.75 0.25 0.75 0.10 0.10 0.27 0.10 0.10 0.10 0.58 0.25 0.24 0.02
449 1.00 1.00 1.00 0.75 1.00 1.00 1.00 1.00 0.72 1.00 1.00 1.00 0.95 0.93 0.88
455 1.00 1.00 1.00 0.75 0.75 0.75 1.00 1.00 0.10 0.80 1.00 1.00 0.69 0.75 0.75
461 0.50 0.30 1.00 0.25 0.10 0.25 0.50 0.10 0.10 0.34 1.00 0.48 0.14 0.34 0.13
462 0.75 0.10 0.50 0.75 1.00 0.50 1.00 0.27 0.10 0.96 1.00 0.20 0.17 0.53 0.63
464 0.10 0.10 0.75 0.25 0.50 0.10 0.10 0.10 0.10 0.10 0.10 0.26 0.08 0.10 0.02
469 0.25 0.10 0.50 0.25 1.00 0.25 0.75 1.00 0.10 0.66 0.81 0.20 0.15 0.52 0.19
474 0.10 1.00 1.00 0.25 0.25 0.10 0.50 0.10 0.10 0.10 0.10 1.00 0.18 0.10 0.09
492 0.75 1.00 1.00 0.50 1.00 0.25 0.75 1.00 1.00 0.26 0.48 1.00 0.79 0.78 0.28
494 0.75 1.00 1.00 0.75 1.00 1.00 1.00 1.00 0.10 0.39 0.76 1.00 0.71 0.61 0.88
504 0.50 1.00 1.00 0.25 0.75 0.50 0.10 0.30 1.00 0.10 0.19 1.00 0.59 0.49 0.04
505 0.50 1.00 1.00 0.25 0.10 0.25 0.50 0.37 0.10 1.00 1.00 1.00 0.40 0.49 0.13
507 0.50 1.00 1.00 0.50 0.75 0.50 0.50 0.27 1.00 0.28 0.52 1.00 0.68 0.54 0.25
511 0.75 0.10 1.00 0.25 0.75 1.00 1.00 1.00 0.10 0.97 0.95 0.33 0.23 0.70 0.63
513 0.50 0.35 0.75 0.25 0.50 0.25 0.50 0.10 0.10 0.10 0.10 0.45 0.14 0.20 0.13
519 1.00 1.00 1.00 1.00 1.00 0.50 1.00 1.00 1.00 0.98 0.90 1.00 0.98 0.99 0.75
525 0.75 0.95 1.00 1.00 0.75 0.75 1.00 0.20 0.50 0.81 1.00 0.96 0.73 0.59 0.88
534 0.50 0.25 0.75 0.25 0.50 1.00 0.75 1.00 0.10 0.10 0.19 0.38 0.12 0.44 0.47
542 0.75 1.00 1.00 1.00 0.75 0.25 1.00 1.00 1.00 0.80 1.00 1.00 0.84 0.91 0.63
547 0.75 0.64 0.75 0.75 0.75 1.00 1.00 0.10 0.50 1.00 1.00 0.67 0.53 0.59 0.88
Appendix II. Continued.
478 WETLANDS, Volume 27, No. 3, 2007