EPA/600/R-92/013
January 1992
A PILOT STUDY TO COMPARE CREATED AND NATURAL WETLANDS
IN WESTERN WASHINGTON AND EVALUATE METHODS
by
Jean C. Sifneos1
Donna L. Frostholm1
Mary E. Kentula2
1ManTech Environmental Technology, Inc.
USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97333
2USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97333
In association with:
Michael Rylko
USEPA Region 10
1200 Sixth Avenue
Seattle, WA 98101
Kathy Kunz
US Army Corps of Engineers, Seattle District
Environmental Resources Section Branch
PO Box C3755
Seattle, WA 98504
Contract Numbers 68-C8-0006 and 6Y0718NAEX
Project Officer:
Eric M. Preston
Wetlands Research Program
USEPA Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97333
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
Accession Number: PBS?143028
u 8 ? 0 S
13:24:13-0 1/30/92
(05)'Waterways Experiment Station, Vicksburg, C 271
MS.*Federal Highway Administration, Washington,
DC. Office of Implementation. S S 350 0 7 00 0
(06) Wetland Evaluation Technique (WET). Volume
2. Methodology. Ope rational Draft. MA 90 0 1S0 0
(09) Final r e p t . J u n 8 4 - S e p 88,
(10) P. R. Adarnus; E. J. Clair aim;
R. D. Smith; R. E. Young (11) Oct 8 7,
(12) 17 1 p
(15) (CN )DTFH61-84-Y-30025
(18)FHWA/IP-S8/029; (18A)
(21) Sponsored by
Federal Highway Administration, Washington, DC.
Office of Implementation.
Message: ENTER: "END" FOR
MENU .
-------
DISCLAIMER
The research described in this report has been funded by the United States
Environmental Protection Agency (EPA) and conducted at EPA's Environmental
Research Laboratory in Corvallis, Oregon, through Contract 68-C8-0006 to ManTech
Environmental Technology, Incorporated, and through Contract 6Y0718NAEX to the
University of Washington. It has been subjected to the Agency's peer and
administrative review, and it has been approved for publication as an EPA document.
Citation: Sifneos, J.C., D.L. Frostholm, and M.E. Kentula. 1991. A Pilot
Study to Compare Created and Natural Wetlands in Western
Washington and Evaluate Methods. EPA/ / - / U.S.
Environmental Research Laboratory, Corvallis, Oregon.
ii
-------
ACKNOWLEDGEMENTS
The authors would like to acknowledge the members of the EPA Wetlands
Research Program and members of the field crews. In particular, we want to
recognize the contributions of personnel from EPA Region 10. Eric Preston, the EPA
Project Officer was supportive of the effort and provided valuable advice. Kathy Kunz
(EPA--currently with the Army Corps of Engineers) and Michael Rylko (EPA)
coordinated all of the site selection, helped with the training session, coordinated the
field crews, and spent long hours in the field. Mary Burg, Washington Department of
Ecology (WDOE); Lynn Childers, U.S. Fish and Wildlife Service (USFWS); Anne Do!d,
Island County Planning; Paul Hickey, Muckleshoot Indian Tribe; Andy McMillan,
WDOE; Mike Nelson, Snohomish County Public Works; Alisa Ralph, US Army Corps
of Engineers; Bill Riley, EPA; Jim Schafer, Washington State Department of
Transportation (WDOT); Dyanne Sheldon, King County Planning; Elaine Somers, EPA;
Doug Swanson, WDOT; Mike Tehan, USFWS; Gary Voerman, EPA; and Bob Zeigler,
Washington State Department of Game assisted in data collection by serving as
members of the field crews. Deborah Coffey performed a quality assurance audit of
the teams in the field and also provided a quality assurance review of the report.
Sheri Confer confirmed the plant species identification. Jane Ely and David Hall
provided data entry. Kristina Miller prepared most of the tables and figures. Jeff
Irish created a draft of the map. Margaret Spence wrote the computer program to
perform the Pielou analysis and Robert Gibson assisted in updating and running it.
Paul Adamus was consulted on aspects of The Wetland Evaluation Technique (WET)
and also reviewed the report.
Also in need of recognition and thanks are Dave Mclntire, Department of Botany
and Plant Pathology, Oregon State University, who assisted with the data analysis and
Linda Haygarth who finalized the map.
Finally, we extend special thanks to those who improved this report through
their review of the draft document. Ann Hairston and Marcia Bollman of ManTech
Environmental Technology, Inc., provided editorial review. Ron Thom and David
Shreffler, of Battelle Pacific Northwest Laboratories, Fred Weinmann of EPA Region
10, Dorothy Milligan of Raedeke Associates Scientific Consulting, and Sandra
Henderson and Paul Shaffer of ManTech Environmental Technology, Inc., reviewed
the draft and offered valuable comments which clarified and strengthened this report.
-------
TABLE OF CONTENTS
DISCLAIMER ii
ACKNOWLEDGEMENTS iii
LIST OF FIGURES vi
LIST OF TABLES vii
ABSTRACT viii
INTRODUCTION 1
METHODS 2
SITE SELECTION 2
DATA COLLECTION 2
QUALITY ASSURANCE AND DATA VERIFICATION 4
DATA ANALYSIS 7
Vegetation data 7
WET data and site description data 8
RESULTS 9
COMPARISON OF CREATED AND NATURAL WETLANDS 9
Freshwater sites 9
Vegetation patterns 9
KOLL 9
NISQ 9
PACF 9
Summary 11
Site description form 11
Pielou 11
WET 14
Saltwater sites 14
Vegetation patterns 14
LINC 14
WILL 14
WEST 17
Summary 17
Site description form 17
Pielou 17
WET 20
iv
-------
COMPARISON OF TEAMS 20
Vegetation Patterns 20
Site description form 23
Pielou 26
WET 26
DISCUSSION 28
COMPARISON OF CREATED AND NATURAL WETLANDS 28
Freshwater wetlands 28
Saltwater wetlands 28
COMPARISON OF TEAMS 28
UTILITY OF METHODS 29
Pielou 29
WET 30
RECOMMENDATIONS 31
CONCLUSIONS 32
LITERATURE CITED 33
APPENDIX I 34
APPENDIX II 38
v
-------
LIST OF FIGURES
Figure 1. The locations of the sites used in the Washington field study ... 3
Figure 2. Dendrograms produced from vegetation data collected at the three
pairs of freshwater wetlands 24
Figure 3. Dendrograms produced from vegetation data collected at the three
pairs of salt marshes 25
vi
-------
/
LIST OF TABLES
Table 1. Summary of the information gathered by teams sampling each of
the nine pairs of created and natural wetlands 5
Table 2. Form used to document the natural features on the site and land
uses in the surrounding area . 6
Table 3a-c. Number of species found on the paired freshwater wetlands and
the number of dominant species categorized by indicator status
and native/introduced 10
Table 4. Summary of site characteristics and surrounding land uses from
the site description form for the freshwater marshes 12
Table 5. Summary of the results of the Pielou (1986) analysis of three pairs
of freshwater marshes 13
Table 6. Ratings of the effectiveness of the three pairs of freshwater
marshes for 12 functions and values as determined by the
Wetland Evaluation Technique (Adamus et al. 1987) 15
Table 7a-c. Number of species found on the paired salt marshes and the
number of dominant species categorized by indicator status and
native/introduced 16
Table 8. Summary of site characteristics and surrounding land uses from
the site description form for the saltwater wetlands 18
Table 9. Summary of the results of the Pielou (1986) analysis of three pairs
of saltwater marshes . . . 19
Table 10. Ratings of the effectiveness of the six pairs of saltwater wetlands
for 12 functions and values as determined by the Wetland
Evaluation Technique (Adamus et al. 1987) 21
Table 11. A comparison of the number of species found by all teams
sampling the created and natural marshes 22
Table 12. The number of differences in the Wetland Evaluation Technique
(Adamus et al. 1987) evaluations for pairs of created and natural
wetlands and for teams visiting the same wetland pair 27
vii
-------
ABSTRACT
Nine created wetlands were paired with nine natural wetlands and compared
for species composition, species diversity, wetland function and other site
characteristics. The sites were located in western Washington and sampled during
July of 1987. The created wetlands selected were mitigation projects for permits
issued under Section 404 of the Clean Water Act (33 U.S.C. A. Section 1344 (1978)).
Each site was sampled by two or four teams comprised of personnel from federal,
state, and local government agencies. Methods used for data collection included a
procedure suggested by Pielou for comparing the diversities of two communities
(Pielou 1986), and the Wetland Evaluation Technique (WET) (Adamus et al. 1987)
which was used to assess potential wetland functions. The goals of the study were
to evaluate the methods used for data collection, compare the created and natural
wetlands, and evaluate the consistency of results obtained by teams sampling the
same site.
Concerns were voiced about the utility of WET and Pielou. The teams had
difficulties answering some of the questions posed by WET. They felt that the
questions were complicated and hard to interpret, or did not make sense for the
wetland type that was being sampled. For the Pielou method, choosing the
appropriate number of species to sample was problematic.
Results from comparisons of species composition and species diversity
indicated that some differences existed between the created and natural sites.
However, because differences also existed in the data collected from different teams
sampling the same site, the created and natural site differences were confounded. In
addition to the possible heterogeneity within the sites, low replicability between the
teams may have been due to insufficient training prior to field work, different botanical
skill levels, or the subjectivity of some of the data collection forms.
A comparison of species composition also found some similarities with respect
to the native/introduced and indicator status of the species (Reed 1988) found at the
created and natural sites. Generally, the species classified as obligate, facultative
wetland, and wetland were native to the Pacific Northwest, while those species
classified as facultative upland and upland were introduced.
viii
-------
SECTION 1
INTRODUCTION
Concern over loss of wetland area and function has accentuated the need to
assess how well created wetlands replace natural wetlands that are destroyed. To
address this issue, the U.S. Environmental Protection Agency's (EPA) Wetland
Research Program (WRP) at the Environmental Research Laboratory in Corvallis (ERL-
C), Oregon, conducted a pilot study in western Washington to determine the
effectiveness of methods that could be used by agencies to evaluate created
wetlands. Additional goals of the study were to compare the characteristics of paired
created and natural wetlands and to compare the results obtained by field crews
sampling the same wetland.
Creating wetlands as compensation for wetland losses is a relatively new
procedure and few studies comparing created and natural wetlands have been
conducted. This lack of scientific research has resulted in limited information on
wetland creation and restoration projects (Kusler and Kentula 1990). The overall
status of the literature on wetland creation and restoration remains uneven by region
and topic. The most quantitative and best documented information is available for
Atlantic coastal wetlands. In addition, most investigations of mitigation are case
studies with no sites included for comparison (Quammen 1986). Consequently, the
WRP implemented studies in four states (Washington, Oregon, Florida, and
Connecticut) to compare natural wetlands with similar wetlands that had been created
as compensatory mitigation under Section 404 of the Clean Water Act (33 U.S.C.A.
Section 1344 (1978)).
Field data were collected in July of 1987 from nine paired created and natural
wetlands located in Washington. Field crews sampled three pairs of freshwater
marshes, three pairs of saltwater marshes, two pairs of mudflats, and one pair of
eelgrass bed. The study tested methods of data collection and the utility of the
methods for use by personnel from resource agencies. The methods employed
relatively inexpensive and simple means of describing and evaluating wetlands, and
attempts were made to use methods that were the least damaging to the study sites.
A vegetation sampling technique suggested by Pielou (1986) was tested and used to
compare the plant communities on each of the paired fresh- and saltwater marshes.
The use of the Wetland Evaluation Technique (WET) (Adamus et al. 1987) to compare
mitigation and reference wetlands was proposed by Adamus (1988) and it was used
to assess the wetland functions on the paired sites. In addition, a form for recording
estimates of the natural features on the wetland and the land uses in the surrounding
area was filled out. The results from WET, the Pielou method, and the site description
form were used to compare the created and natural wetlands, and to compare teams
sampling the same wetland.
1
-------
SECTION 2
METHODS
SITE SELECTION
Nine created wetlands were paired with nine natural wetlands and compared
for species composition, species diversity, wetland function, and other site
characteristics (e.g., natural features of the wetland and the surrounding area). The
sites were sampled during July of 1987 by either two or four teams comprised of four
individuals each. Time constraints, and the desire to decrease the amount of
trampling and degradation to the sites by sampling, prevented all teams from visiting
all sites as was originally planned.
The created wetlands were created as functional replacement for wetlands lost
under Section 404. They were chosen from the Section 404 permit files of the
Seattle District of the U.S. Army Corps of Engineers (COE) and the Regional Office
(Region 10) of the EPA. Four character codes (e.g., KOLL) were used as pseudonyms
for the permit numbers to preserve the anonymity of the sites. All created wetland
projects which had been completed by the summer of 1987 were included in this
study. The wetlands ranged in age from 7-28 months and in size from 0.1 to 5.3
acres (0.4 to 2.1 ha). They represented four wetland types: freshwater marsh, salt
marsh, mud flat, and eelgrass bed (Zostera marina L.) (Appendix I). The criteria used
for pairing the created and natural wetlands were type, size, and proximity. The
regulatory personnel involved in permitting assisted in finding natural wetlands as
similar as possible to the destroyed wetland. Four of the site pairs were adjacent to
each other. For these sites, natural wetlands were remnants of the wetlands
destroyed (i.e., the wetlands for which the created wetlands were to compensate).
The rest of the natural wetlands were located from 1-30 miles (0.6-18 km) from the
created sites (Appendix I). Six of the paired wetlands were located in the Puget
Sound area, one was located outside of Olympia, and two were located in Gray's
Harbor (Figure 1).
DATA COLLECTION
The study was a cooperative effort with EPA Region 10 and other federal, state
and local agencies in the area. The teams were comprised of personnel from these
agencies and individual members varied to a certain extent in their levels of field
experience. The teams performed a WET evaluation of the site, filled out a site
2
-------
Washington Field Study Sites
will
WEST
KOLL
BLR1
LINC
EACH DOT REPRESENTS
PAIRED STUDY SITES
(not to scale)
SEATTLE TACOMA
METROPOLITAN AREA
Washington
Figure 1. The locations of the sites used in the Washington field study.
3
-------
description form, and collected vegetation data necessary for comparing the paired
sites using a method devised by Pielou (1986) (Table 1). The sites were sampled by
more than one team so that their results could be compared and the comparability of
the methods ascertained.
WET was used to assess the wetland's functions and values. The technique
assigns ratings of high, medium, and low to the following wetland functions and
values: groundwater recharge, groundwater discharge, floodflow alteration, sediment
stabilization, sediment/toxicant retention, nutrient removal/transformation, production
export, aquatic diversity/abundance (d/a), habitat suitability for fish and wildlife,
wildlife breeding, migration and wintering, uniqueness/heritage, and active recreation.
The ratings are assigned to the functions and values in terms of three different
categories: significance (value to society), opportunity (whether a wetland has the
opportunity to perform a function or value), and effectiveness (the probability of a
wetland being able to maximize the opportunity to perform a function or value)
(Adamus et al. 1987). To perform the evaluations, the teams collected map,
background, and field data on the wetlands and answered the questions on the WET
forms in the office and field.
The site description form consisted of 11 questions quantifying the percent of
natural features and land uses on the site and in the surrounding area (Table 2). The
percent and type of disturbance were also recorded.
The Pielou method was used to compare the diversities of six pairs of wetlands.
The mudflats were excluded because they were not vegetated, and the eelgrass bed
was excluded because it was a monotype. One to six transects were established at
each wetland to sample the vegetation. Rectangular quadrats (1 m2) were placed at
regular intervals along the transects and the six species closest to the center of the
quadrat were recorded. Forty quadrats were sampled per site. The transects were
located to best typify the vegetation communities on the site. If an environmental
gradient was present (e.g., an elevation gradient), the transects were placed parallel
to the gradient to representatively sample the plant communities.
QUALITY ASSURANCE AND DATA VERIFICATION
To increase the likelihood of accurate and comparable data collection, the teams
were trained in field protocols prior to field work. Representatives from EPA and COE
held a one week training session in Olympia, Washington, to explain the methods for
data collection and to provide an opportunity to practice filling out WET forms in the
field.
During the field season, a quality assurance audit was performed to evaluate
the performance of the four field crews. The auditor checked the field crews to
4
-------
Table 1.
Summary of the information gathered by teams sampling each of the nine pairs
of created and natural wetlands.
Ln
Wetland type
Fres
hwater m
arsh
Sal
twater m
arsh
Eelgrass
bed
Mud
flat
Site acronym
KOLL
NISQ
PACF
LINC
WEST
WILL
SEQU
BLRI
PIER
Number of teams
sampling wetland
4
2
2
4
2
2
2
4
4
WET evaluation
performed?
yes
yes
yes
yes
yes
yes
yes
yes
yes
Pielou/vegetation
patterns analysis
performed?
yes
yes
yes
yes
yes
yes
no
no
no
Site descriptions
form filled out?
yes
yes
yes
yes
yes
yes
yes
by 1
of 4
teams
by 1
of 4
teams
-------
Table 2. Form used to document the natural features on the site and land uses
in the surrounding area.
FORM A. Qualitative site information [to be used with sketch map]
Site Name :
Personnel Date
Mappable wetland characteristics: Sketch or label the following items on the
base map. Indicate associated percentage values on this form where requested.
1. Indicate north.
2. Access point.
3. Hydrologic features: (a) locations of inlet/outlet, (b) major channels {where
applicable), (c) direction of water flow, (d) obstructions to water flow.
4. % open water : vegetation
5. % wetland inundated
6. % wetland that is disturbed
7. Vegetation zonation or patches
8. Label dominant vegetation types, indicate % relative cover for:
(a) % trees
(b) % shrubs
(c) % emergent herbs
(d) % submergent herbs
(e) % nonvegetated area (natural)
(f) % nonvegetated area (disturbance related)
9. Label surrounding area, indicate % relative cover.
(a) % forest
(b) % meadow/field
(c) % shrubs
(d) % human disturbance
(1) % cultivation
(2) % industrial, specify
(3) % housing
(4) % highway
(5) % grazing
(6) % commercial
*1-6 should total the precentage value in (d)
10. Draw in transects on the sketch map. Indicate the length and direction of each
from its origin.
11. Comments: Other pertinent site information may be written on the back of
this form.
6
-------
determine whether they followed protocols and directions thoroughly and collected
the data in the correct order.
t
The botanists on each team identified the vegetation species in the field and
specimens were brought back to the EPA's ERL-C to be verified. All specimens were
archived at the Laboratory.
To ensure that errors did not occur in transferring the data from the field sheets
to computer files, field data were double entered by two individuals working
independently. After entry, the two data sets were electronically compared.
Discrepancies were corrected by comparison with the field sheets until both data sets
were in exact correspondence.
DATA ANALYSIS
Vegetation data
The vegetation data were compiled and analyzed using a method suggested by
Pielou (1986). The method compared the most typical quadrat (i.e., the quadrat most
similar to the other quadrats sampled on a site) from each of a pair of created and
natural wetlands. Two tests were performed. The null hypothesis of the first test
was that the created and natural sites had similar diversities. If the null hypothesis
was rejected, then the second hypothesis tested that the plant community associated
with the created wetland was a subset of the plant community of the natural wetland.
Alternatively, if the null hypothesis of the first test was not rejected, the second
hypothesis tested was that communities from the created and natural wetlands were
from the same parent population. The analysis was used to compare the site pairs
and the results obtained by different teams sampling the same sites.
The vegetation data were also used to summarize and compare the species
composition at the created and natural sites. The plants which could not be
identified to species were included in the analysis. Species that were found only once
at any paired site were deleted from the analysis because they were considered rare.
The total number of times that each species was found by each team at each site was
tallied. This included the number of species found only on the created site, the
number of species found only on the natural site, and the number of species found on
both members of a site pair. Thus, comparisons were made between the six paired
sites, between the three paired freshwater sites, and between the three paired
saltwater sites.
The dominant species on each site were compared to see if the species
characterizing the created wetlands were similar to those characterizing the natural
wetlands. Dominance was determined by the total number of times a species was
7
-------
found at each site by all teams sampling the site. The species found the greatest
number of times (i.e., those species found in the quadrats most often) were
considered dominant. This procedure inherently involved some degree of subjectivity,
however, in all cases, similar methods of determining dominant species were used.
The National List of Plant Species that Occur in Wetlands: Northwest (Region
9) (Reed 1988) was used to determine the indicator status, and Hitchcock and
Cronquist (1981) was used to determine whether the species were native or
introduced. The indicator categories are:
o obligate wetland-estimated 99% probability of occurring in wetlands
o facultative wetland-estimated 67-99% probability of occurring in
wetlands
o facultative-estimated 34-66% probability of occurring in wetlands
o facultative upland-estimated 34-66% probability of occurring in
nonwetlands
o obligate upland-may occur in wetland in another region, but 99%
estimated probability of occurring in nonwetlands.
For the purposes of this report, those species which were in the obligate or
facultative wetland categories were considered to be wetland species, those species
in the facultative category were considered to be facultative species, and those
species in the facultative and obligate upland categories were considered to be upland
species. A list of the species found in the field study and their indicator and
native/introduced status can be found in Appendix II.
Finally, the data were also used to perform a cluster analysis for each paired
site to determine whether the plant community on the created site differed from that
on the natural site, and if the answer depended on which team did the sampling.
WET data and site description data
The ratings obtained from the WET assessment for the different functions and
values were compared for the pairs of sites and for the different teams which sampled
the same site. Only the ratings from the effectiveness category were compared
because it was considered the most objective category (P. Adamus, ManTech
Environmental Technology Inc., Corvallis, OR, pers. comm.). Similarly, the results
from the analysis of the site description data compared the percentages on the forms
for each pair of sites and for the different teams sampling the same site.
8
-------
SECTION 3
RESULTS
COMPARISON OF CREATED AND NATURAL WETLANDS
The following section compares the vegetation found on the created and natural
wetlands by wetland pairs. The pairs are grouped into salt- and freshwater marshes.
Results from the site description form, Pielou (1986), and WET (Adamus et al. 1987)
are also described for the fresh- and saltwater wetlands.
Freshwater sites
Vegetation patterns-
KOLL—Eighty-seven species were found by the four teams that sampled the
KOLL paired wetlands: 22 species were unique to the created wetland, 33 species
were unique to the natural wetland, and 32 species were found on both wetlands
(Table 3a). The percent of wetland and facultative species found on the created and
natural sites was similar: 61 % for the created and 58% for the natural. The wetland
and facultative species were predominately native to the Pacific Northwest rather than
introduced. Ten of the 54 species found on the created site, and 12 of the 65
species found on the natural site, were classified as dominant. Six of the ten
dominant species on the created site were introduced however, most (9/12) of the
dominant species found on the natural site were native.
NISQ--Fiftv-two species were found by the two teams that sampled the NISQ
paired wetlands (Table 3b). The majority of the species were found on either the
created or natural wetland: 42% of the species were unique to the created site and
38% of the species were unique to the natural site. Ten species (19%) were found
on both of the sites. The combined percents of wetland and facultative species
indicated a difference between the two wetlands: 44% of the species on the created
site were wetland and facultative compared with 63% on the natural site. Most of
the wetland and facultative species were native (79%), while the upland species were
primarily introduced (69%). Eight of the 32 species found on the created site and
seven of the 30 species found on the natural site were classified as dominant. Four
of the eight dominant species found on the created site were both upland and
introduced species.
PACF—Fifty species were found by the two teams that sampled the PACF
paired wetlands (Table 3c). Of these, 1 2 species were found only on the created site,
19 species were found only on the natural site, and 19 species were found on both
sites. The combined percent of wetland and facultative species found on the created
9
-------
TABLE 3a-c. Nuiter of species foind on the paired freshwater wetlands and the nuttier of
dominant species categorized by indicator status (]NO. STAT.) and native/introduced (N/I) (Reed
1988). NAT is the nutter of species fowd only on the natural site, CR is the nurber of
species fokwid only on the created site and BOTH is the nuiter of species, found on both paired
sites. Dashes (--) indicate that the species was not fo^rtd at the site(s). Nunbers in
parentheses after NAT and CR indicate the total muter of species found at the natural and
created sites, respectively. The nutbers in parentheses following the TOTALS are the percent
of the total nunber of species or the dominant species, fornd at the natural, created, or both
sites. 06L is an obligate wetland species; FACU is facultative wetland species; FAC is a
facultative species; FACU is a facultative ipland species; UPL is an inland species; N is a
native species; I is an introduced species; UNKNOWN means the indicator status could not be
determined for the plant.
Table 3a. KOLL
INO. STAT.
ALL SPECIES (87)
DOMINANT
SPECIES (19)
& N/1
NA
(65) BOTH
CR<54>
NAT(12> BOTH
CR(IO)
OBL/N
i
1
--
--
OBL/I
1
--
--
--
--
FACV/N
11
1
1
--
3
FACV/I
-
6
1
--
2
1
FAC/N
2
3
1
--
FAC/1
-
2
3
--
--
--
FACU/M
2
--
3
--
--
FACU/I
S
6
1
--
3
UPL/N
--
--
--
--
--
UPL/1
--
2
--
--
--
UNKNOWN
1
5
--
--
TOTALS
3
(38*) 32
(37X) 22 (25%) 9 (47*)
3
(16%) 7 (37X)
TABLE 3b. N1SO
]ND. STAT.
& N/I
ALL SPECIES
NAT(30) BOTH
(52)
CR(32)
| DOMINANT SPECIES (14)
! NAT(7) BOTH CR(8)
OBL/N
U
1
2
; 2
--
--
06L/I
1
--
--
i
--
•-
FACU/N
(
3
4
| 1
1
--
FACV/I
1
--
1
I
--
--
FAC/N
2
1
1
i
--
2
FAC/1
2
--
1
| 2
--
--
FACU/N
3
--
1
1
--
--
FACU/I
1
2
3
J 1
--
2
UPL/N
--
--
1
1
--
--
UPL/1
--
2
3
1
--
2
UNKNOWN
2
1
5
1
--
1
TOTALS 20 ( 38*) 10 (19X) 22 (42X) 6 (43X) 1 (7X) 7 (50X)
TABLE 3c. PACF
IND. STAT.
ALL
SPECIES
(50)
11
11
DOMINANT SPECIES (15)
i N/I
NAT(38) BOTH
CR(31)
i i
NAT(fl) BOTH CRC7)
OBL/N
1
3
1
11
2 -- 2
OBL/I
2
--
J i
1
FACW/N
5
8
6
!!
2 --4
FACV/I
2
1
¦ i
¦ i
1
FAC/N
A
3
1
¦ i
¦ i
2 -- 1
FAC/1
1
-¦
--
¦ i
¦ i
¦ i
11
--
FACU/N
2
--
--
FACU/I
3
--
1
11
* i
--
UPL/N
--
--
11
11
--
UPL/I
1
--
1
11
11
..
UNKNOWN
1
1
11
TOTALS
19 (38*)
19 (38X) 12 (24X)
8 (53*) -- (OX) 7 (47X)
10
-------
and natural wetlands were similar; 87% for the created site and 82% for the natural
site. The wetland and facultative species also had a greater number of native than
introduced species. Seven of the 31 species found on the created wetland and eight
of the 38 species found on the natural wetland were classified as dominant. All of
the dominant species on the created and natural sites were facultative or wetland
species, and all of the species found on the created site were native to the Pacific
Northwest.
Summary-The number of species found at the three paired wetlands varied as
did the percentage of wetland and facultative species on the sites. However,
similarities were also found between the wetlands. Most of the wetland and
facultative species were native, and the majority of upland species were introduced.
The percent of species classified as dominant ranged from 18-25% for each site. The
majority of the dominant species were found on either the created or the natural site;
few, if any, dominant species were found on both of the paired sites.
Site description form-
Comparisons were made of the results from the same team evaluating the
paired created and natural wetlands for three questions of interest on the site
description form: percent of site covered by open water, percent of site that was
disturbed, and percent of surrounding area that was disturbed (Table 4). There were
eight comparisons for each of the three questions (i.e., the three freshwater sites
were visited by more than one team each). Differences between the paired sites were
indicated by answers which differed by at least 40%. Two of the eight comparisons
(25%) differed for the percent of the site covered by open water. In both cases the
created wetland had more open water than the natural wetland. Three of the
comparisons (38%) differed for the percent of the site that was disturbed. In all three
cases, the created wetland was more disturbed than the natural wetland. Three of
the comparisons (38%) also differed for the percent of the surrounding area that was
disturbed and again, in all three cases, the created wetland was more disturbed than
the natural wetland.
Pielou—
The results from the Pielou evaluations did not always agree for the three pairs
of freshwater wetlands (Table 5). At the KOLL sites, the results from three of the
four teams rejected both of the hypotheses associated with test 1 and test 2
concluding that the plant community on the created site was less diverse, and not a
subset of the plant community on the natural site. For the NISQ sites, the results
from one team suggested that the two communities had similar diversities but were
not from the same parent population, while the results from the other team suggested
that the plant community on the created site was less diverse, and not a subset of the
community on the natural site. The evaluations of the two teams agreed that the
11
-------
Table 4. Summary of site characteristics and surrounding land uses from the site
description form for the freshwater marshes. C = created, N = natural, OW = % open
water, DIST =% wetland disturbed, SUR DIST = % surrounding area that is disturbed.
SUR
SITE
C/N
TEAM
OW
DIST
DIST
KOLL
C
1
10
80
60
KOLL
N
1
10
10
60
KOLL
C
2
10
100
15
KOLL
N
2
5
40
50
KOLL
C
3
10
0
85
KOLL
N
3
10
5
40
KOLL
C
4
10
100
45
KOLL
N
4
10
100
50
NISQ
C
2
60
50
75
NISQ
N
2
5
0
30
NISQ
C
4
50
50
50
NISQ
N
4
1
15
70
PACF
C
1
5
10
50
PACF
N
1
5
5
20
PACF
C
3
30
0
70
PACF
N
3
15
25
30
12
-------
Table 5. Summary of the results of the Pielou (1986) analysis of three pairs of
freshwater marshes sampled by two or four teams. For test one, values of Z
> 1.28 (one tailed test, alpha =0.10) caused the hypothesis that the two
communities had the same diversity to be rejected, and supported the conclusion
that the created site was less diverse than the natural site. If test one was
not rejected, then values of Z > 1.282 for test two caused the hypothesis that
the created community was a subset of the natural community to be rejected.
Alternately, if test one was not rejected, then values of Z >±_1.65 for test
two (two tailed test, alpha =0.10) caused the hypothesis that the created and
natural communities were from the same parent population to be rejected.
* = rejected.
TEST 1 TEST 2
SITE TEAM Z VALUE Z VALUE IMPLICATIONS
KOLL
1
•
3.56
•
4.36
The created site is less diverse than the natural site.
The created site is not a subset of the natural site.
2
0.39
•
3.08
The two communities have similar diversities
but are not from the same parent population.
3
•
2.79
•
4.99
The created site is less diverse than the natural site.
The created site is not a subset of the natural site.
4
•
2.95
•
2.58
The created site is less diverse than the natural site.
The created site is not a subset of the natural site.
NISQ
2
0.98
•
2.26
The two communities have similar diversities
but are not from the same parent population.
4
•
1.82
2.31
The created site is less diverse than the natural site.
The created site is not a subset of the natural site.
PACF
1
-0.01
•
8.59
The two communities have similar diversities
but are not from the same parent population.
3
0.90
•
10.00
The two communities have similar diversities
but are not from the same parent population.
13
-------
V.
plant communities on the paired PACF wetlands had similar diversities but were not
from the same parent population.
WET-
WET evaluations were also compared for the three pairs of freshwater wetlands
(Table 6). At the KOLL sites, the created and natural wetlands were rated similarly
for many of the functions, however, there did seem to be a difference between the
paired sites for floodflow alteration, sediment/toxicant retention, and general fish
habitat. The created and natural sites were rated exactly the same at the NISQ sites
for all but sediment stabilization and general fish habitat. At the PACF sites, the
created and natural wetlands were rated similarly by the teams for all but
sediment/toxicant retention and wildlife (d/a) wintering. In general, all sites were low
for groundwater recharge, aquatic d/a and wildlife d/a breeding, and medium for
production export.
Saltwater sites
Vegetation patterns--
LINC-Fiftv-seven species were found by the four teams that sampled the UNC
paired wetlands: 21 species were unique to the created site, 23 species were unique
to the natural site, and 13 species were found on both sites (Table 7a). The percent
of wetland and facultative species was not similar for the paired wetlands: 71 % were
found on the created site and 56% were found on the natural site. This difference
may be partially attributed to the percentage of species (28%) which could not be
identified. Most wetland and facultative species were native to the Pacific Northwest,
however, most upland species were introduced. Seven of the 34 species on the
created site and seven of the 36 species on the natural site were classified as
dominant. All of the dominants were wetland species except for one unknown. Also,
all of the dominant species were native except for one introduced species on the
natural site.
WILL-Thirtv-seven species were found by the two teams that sampled the
WILL paired wetlands (Table 7b). Nearly half (43%) of the species were found on
both sites. The created and natural sites had a total of 26 and 27 species,
respectively. The combined percent of wetland and facultative species was similar
for the two wetlands: 81 % for the created and 74% for the natural. Most (79%) of
the wetland and facultative species were native. Four of the 26 species on the
created wetland and three of the 27 species on the natural wetland were classified
as dominant. All of the dominant species were also wetland species except for one
upland species found on the natural site.
14
-------
Table 6. Ratings [high (H), medium (M), low |L), unknown (U)] of the effectiveness of
the three pairs of freshwater marshes for 12 functions and values as determined
by the Wetland Evaluation Technique (WET) (Adamus et al. 1987). The WET
evaluations were generated by two or four teams per site during the summer of
1987 in Washington. C=created, N = natural, GWR = groundwater recharge;
GWD = groundwater discharge; FFA = floodflow alteration; SS = sediment stabilization;
S/TR = sediment/toxicant rentention; NR/T = nutrient removal/transformation;
PE = production export; AD/A aquatic diversity/abundance; GFH = general fish
habitat; WB = wildlife diversity/abundance breeding; WM = wildlife diversity/
abundance migration; WW = wildlife diversity/abundance wintering.
SITE TEAM C/N GWR GWD FFA SS S/TR NR/T PE AD/A GFH WB WM WW
KOLL
1
C
M
L
H
H
L
M
L
M
H
H
N
M
M
H
L
L
M
L
M
H
H
2
C
M
L
H
H
L
M
L
M
H
H
N
M
M
H
L
L
M
M
L
H
H
3
C
M
L
H
H
L
M
M
H
H
N
M
M
M
L
M
L
L
L
4
C
H
L
M
L
L
M
L
M
H
H
N
M
M
H
L
L
M
L
L
H
M
NISQ
2
C
L
L
H
H
H
M
L
M
M
M
N
L
L
H
H
H
M
L
M
M
M
4
C
L
L
M
H
H
M
L
M
M
M
N
L
L
H
H
H
M
L
L
M
M
PACF
1
C
M
L
H
L
L
M
L
M
H
M
N
M
L
H
H
L
M
L
M
H
H
3
C
M
L
H
L
L
M
L
M
M
M
N
M
L
L
H
H
M
L
M
H
H
-------
TABLE 7a-c. Nuitoer of species found on the paired salt marshes and the r*jit>er of dominant
species categorized by indicator status (IND. STAT.) and native/introduced (N/1) (Reed 19B8).
NAT is the r*jit>er of species fotxid only on the natural site, CR is the nuifcer of species found
only on the created site and BOTH is the nurber of species found on both paired sites. Dashes
(--) indicate that the species Mas not focnd at the site(s). Nurbers in parentheses after MAT
and CR indicate the total ninber of species fo^nd at the natural and created sites,
respectively. The ntnijers in parentheses following the TOTALS are the percent of the total
nurfcer of species or the dominant species, found at the natural, created, or both sites. 06L
is an obligate wetland species; FACW is facultative wetland species; FAC is a facultative
species; FACU is • facultative ipland species; UPL is an upland species; N is a native
species; 1 is an introduced species; UNKNOUN means the indicator status could not be determined
for the plant.
Table 7a. LINC
IND. STAT. ALL SPECIES (57)
& N/I NAT(36) BOTH CR(34)
j| DOMINANT SPECIES (14)
!! NAT(7) BOTH CR(7)
08L/N 4 4 7
06L/I -- -- 1
FACW/N 3 3 4
FACU/1 1 1 3
FAC/N 2
FAC/1 1 1
FACU/N 1
FACU/1 2 1"
UPL/N
UPL/1 2
UNKNOUN 7 3 6
!! 2 -5
!! 4 --1
!! i
j j
TOTALS 23 (41X) 13 (23X) 21 (36X) 7 (50X) -- (OX) 7 (50X)
TABLE 7b. UILL
IND. STAT. ALL SPECIES (37) j
t N/I NAT(27) BOTH CRC26) !
DOMINANT SPECIES (7)
NAT(3) BOTH CR<4)
06L/N 4 5 3 !
06L/I — 1 1 j
FACW/N -- 5 3 |
FACW/1 1 2 1 [
FAC/N 2 J
FAC/I -- -- — 1
FACU/N 2 " j
FACU/1 -- 1 1 |
UPL/N 1 -- -- J
UPL/I -- -- " !
UNKNOUN 1 2 1!
2 " 1
1
2
1
TOTALS 11 (30X) 16 (43X) 10 (27X) 3 (43X) -- (OX) 4 (57X)
TABLE 7c. WEST
IND. STAT. ALL SPECIES (27) |
& N/I NAT(22) BOTH CR(23) !
DOMINANT SPECIES (8)
NAT(4) BOTH CR(5)
OBl/N 1 6 — |
OBL/I 1 |
FACW/N -- 9 2 i
FACW/I -- 3 — {
FAC/N -- -- -• |
FAC/1 -- -- -- |
FACU/N -- — i
FACU/1 — -- 1 |
UPL/N 1 -- 1 |
UPL/I " " " j
UNKNOUN 1 1 !
3-2
1 2
TOTALS 4 (15X) 18 (67*) 5 (27X) 3 (38X) 1 (13X) 4 (50X)
16
-------
WEST—Twenty-seven species were found by the two teams that sampled the
WEST paired wetlands (Table 7c). Five species were found on the created site, four
species were found on the natural site, and 18 species were found on both. The
percent of wetland and facultative species on the paired wetlands was similar: 87%
were found on the created and 91% were found on the natural. More native than
introduced wetland species were found. Of the 23 species found on the created site,
five were classified as dominant and of the 22 species found on the natural site, four
were classified as dominant. All of the dominant species were also wetland species.
Summary—The results indicated differences between the paired wetlands. The
percent of the species found on both the created and natural sites ranged from 23-
67% and the percent of the species classified as dominant ranged from 11-22%.
However, similarities were also found. For example, all dominant species were also
wetland species, except for the WILL natural site which had one dominant species
that was upland. In general, the dominant species were found on either the created
or the natural sites.
Site description form-
Comparisons were made of the results of the same team evaluating the paired
created and natural wetlands for the three questions of interest on the site description
form (Table 8). There were 12 comparisons for each of the three questions.
Differences between the created and natural saltwater wetlands were indicated by
answers which differed by at least 40%. Eight percent of the comparisons were
different for the percent of the site covered by open water. For this comparison, the
created wetland had more open water than the natural wetland. Thirty three percent
of the comparisons differed for the percent of the site that was disturbed. In three
of the comparisons, the created wetland was more disturbed than the natural wetland,
and in one of the comparisons, the natural wetland was more disturbed. One of the
12 comparisons differed for the percent of the surrounding area that was disturbed.
For this comparison, the created wetland had a higher percent of disturbance in the
surrounding area than the natural wetland.
Pielou-
The results from the Pielou evaluations performed by both of the teams that
sampled the WEST and WILL paired wetlands agreed that the plant communities on
the paired wetlands had similar diversities, but were not from the same parent
population (Table 9). However, the team's evaluations were not in agreement for the
LINC paired wetland. The evaluations from two of the teams concluded that the two
communities had similar diversities but were not from the same parent population,
while the evaluations from the other two teams suggested that the community on the
created site was less diverse, but not a subset of the community on the natural site.
17
-------
Table 8. Summary of site characteristics and surrounding land uses from the site
description form for the saltwater wetlands. C = created, N = natural, 0W = % open
water, DIST = % wetland disturbed, SUR DIST = % surrounding area that is disturbed.
SUR
SITE
C/N
TEAM
OW
DIST
DIST
LINC
C
1
50
5
80
LINC
N
1
50
10
0
LINC
C
2
60
0
100
LINC
N
2
0
100
90
LINC
C
3
35
0
75
LINC
N
3
50
0
55
LINC
C
4
70
0
30
LINC
N
4
60
0
50
WEST
C
1
10
70
50
WEST
N
1
20
10
50
WEST
C
2
10
0
30
WEST
N
2
10
0
30
WILL
C
1
10
75
0
WILL
N
1
20
10
10
WILL
C
2
30
0
10
WILL
N
2
10
0
10
SEQU
C
3
100
0
25
SEQU
N
3
75
0
40
SEQU
C
4
75
5
5
SEQU
N
4
66
0
10
BLRI
C
3
90
100
100
BLRI
N
3
85
15
80
PIER
C
3
100
0
100
PIER
N
3
100
0
70
18
-------
Table 9. Summary of The results of the Pielou (1986) analysis of three pairs of
saltwater marshes sampled by two or four teams. For test one, values of Z
> 1.28 (one tailed test, alpha = 0.10) caused the hypothesis that the two
communities had the same diversity to be rejected, and supported the conclusion
that the created site was less diverse than the natural site. If test one was
not rejected, then values of Z > 1.282 for test two caused the hypothesis that
the created community was a subset of the natural community to be rejected.
Alternately, if test one was not rejected, then values of Z >±_1.65 for test
two (two tailed test, alpha =0.10) caused the hypothesis that the created and
natural communities were from the same parent population to be rejected.
* = rejected.
TEST 1 TEST 2
SITE TEAM Z VALUE Z VALUE IMPLICATIONS
LINC
1
«
2.55
•
4.94
The created site is less diverse than the natural site.
The created site is not a subset of the natural site.
2
0.48
«
4.56
The two communities have similar diversities
but are not from the same parent population.
3
•
1.55
•
4.66
The created site is less diverse than the natural site.
The created site is not a subset of the natural site.
4
1.27
•
6.27
The two communities have similar diversities
but are not from the same parent population.
WEST
1
-0.86
•
8.06
The two communities have similar diversities
but are not from the same parent population.
2
1.18
«
7.12
The two communities have similar diversities
but are not from the same parent population.
WILL
1
-2.98
«
2.37
The two communities have similar diversities
but are not from the same parent population.
2
-1.91
•
5.39
The two communities have similar diversities
but are not from the same parent population.
19
-------
WET-
Similarities between the created and natural sites were found in many of the
WET evaluations (Table 10). In general, all of the evaluations of the saltwater sites
(i.e., salt marshes, mudflats and eelgrass beds) (>97%) had low ratings for
groundwater recharge, floodflow alteration, sediment/toxicant retention, nutrient
removal/transformation, and wildlife d/a breeding. Many of the evaluations of the
sites (>83%) were also low for groundwater discharge. More than 88% of the
evaluations had a medium rating for production export and general fisheries habitat.
The four remaining functions, sediment stabilization, aquatic d/a, wildlife d/a migration
and wildlife d/a wintering, varied by team, created or natural site pair, site within a
wetland type, and wetland type. The salt marshes had primarily mediums and highs
for these functions. The eelgrass sites were low for aquatic d/a, primarily high for
wildlife d/a migration and varied for wildlife d/a wintering and sediment stabilization.
The mudflats were medium for aquatic d/a, but varied by created and natural site for
the other functions.
COMPARISON OF TEAMS
Vegetation Patterns
Two paired wetlands (KOLL and LINC) were sampled by all four teams.
Comparisons between these sites showed that variability existed among the teams
with respect to the number of times a species was found on a site. The comparisons
showed that 35-54% of the total number of plant species recorded were found by
only one of the teams (i.e., the other three teams did not find the species).
Using only the species that the teams agreed were on the sites (for KOLL and
LINC, three of the four teams had to have found the species; for the remaining four
pairs, the species had to be found by both teams six or more times), coefficients of
variation (CV) were calculated to determine how similar each team's findings were.
The results indicated that variability existed between the number of times each
species was found by each team at each site. The average CV for each site ranged
from 16-74 (Table 11). Three wetlands had an average CV of 27 or less, the
remaining nine had an average CV of 45 or more. The NISQ sites had similar between
team findings as indicated by a CV of 25 or less for 80% and 86% of the species
found on the created and natural sites, respectively. The percent of species having
a CV of 25 or less at the remaining sites ranged from 6-69%.
The average CV for the freshwater wetlands was 56 for the created sites and
41 for the natural sites. The average CV for the saltwater wetlands was similar: 51
for the created sites and 50 for the natural sites.
20
-------
Table 10. Ratings [high (H), medium (M), low (L), unknown (U)] of the effectiveness of
the six pairs of saltwater wetlands for 12 functions and values as determined
by the Wetland Evaluation Technique (WET) (Adamus et al. 1987). The WET
evaluations were generated by two or four teams per site during the summer of
1987 in Washington. C = created, N = natural, GWR = groundwater recharge;
GWD = groundwater discharge; FFA = floodflow alteration; SS = sediment stabilization;
S/TR = sediment/toxicant rentention; NR/T = nutrient removal/transformation;
PE = production export; AD/A aquatic diversity/abundance; GFH = general fish
habitat; WB = wildlife diversity/abundance breeding; WM = wildlife diversity/
abundance migration; WW = wildlife diversity/abundance wintering.
SITE TEAM C/N GWR GWD FFA SS S/TR NR/T PE AD/A GFH WB WM WW
LINC 1
C L
L L H
L
M
H
M
H
H
N L
L L H
L
U
H
M
H
M
2
C L
L L M
L
M
H
M
H
H
N L
L L H
L
M
M
M
H
M
3
C L
L L M
L
M
H
M
H
H
N L
L L H
L
M
M
M
M
M
4
C L
L L H
L
U
H
M
M
M
N L
L L H
L
M
H
M
H
H
WEST 1
C L
L L H
L
M
M
M
H
M
N L
L L H
H
M
M
M
H
M
2
C L
L L M
L
U
M
M
H
H
N L
L L M
L
U
M
M
H
H
-------
Table 10. continued
SITE TEAM C/N GWR GWD FFA SS S/TR NR/T PE AD/A GFH WB WM WW
WILL
1
C
L
H L
L M
H
M
H
H
N
M
H L
L M
M
M
H
H
2
C
M
H L
L M
M
M
H
H
N
M
H L
L M
M
M
H
H
SEQU
3
G
L
H L
L M
L
M
H
M
N
L
M L
L M
L
M
H
H
4
C
L
H L
M M
L
M
L
L
N
M
L L
L M
L
L
H
H
BLR1
1
C
L
L L
L M
M
M
L
L
N
H L
L M
M
M
H
M
2
C
L
L L
L M
M
M
M
M
N
L
H L
L M
M
M
H
M
3
C
L
L L
L M
M
M
M
M
N
L
H L
L M
M
M
H
M
4
C
L
L L
L M
M
M
L
L
N
L
H L
L M
M
L
H
M
PIER
1
C
L
L L
L M
M
M
L
L
N
M
L L
L M
M
M
L
L
2
C
L
L L
L M
M
M
M
M
N
M
L L
L M
M
M
L
L
3
C
L
L L
L M
M
M
L
L
N
L
L L
L M
M
M
L
L
4
C
L
H L
L M
M
M
L
L
N
H L
L M
M
M
L
L
-------
Table 11. A comparison of the nunber of species found by alt teams sampling the created (C) and natural (N) marshes. CV is
the coefficient of variation.
No. of species
found on wetlands
Mo. of species having
a CV less than 25.0
X of species having
a CV less than 25.0
Average CV for wetland
KOLL
N C
2 4 17
2 1
6 6
72 74
FRESHWATER
NISQ
N C
15 7
12 6
80 66
18 16
PACF
N C
16 12
11 5
69 42
47 54
LINC
N C
12 7
1 2
8 29
66 48
SALTWATER
WILL
N C
7 11
4 2
57 18
27 45
WEST
N C
9 11
2 3
22 27
66 59
22
-------
A cluster analysis was performed on the six paired wetlands. The dendrograms
from the analysis show that the different teams sampling the same site consistently
clustered before the paired sites (Figures 2 and 3). This suggests that the species
found by different teams at the same site were more homogeneous or similar than the
species found on the paired created or natural sites.
Site description form
Comparisons were made of the results obtained by different teams visiting the
same site (Table 4 and Table 8). There were seven pairs of sites evaluated by two
or four teams (the remaining two site pairs were sampled by one team each). Again,
differences greater than or equal to 40% were noted for the percent of the site
covered by open water, percent of the site that was disturbed, and the percent of the
surrounding area that was disturbed. One of the 14 sites (7%) had differences of at
least 40% for the percent of the site covered by open water, five sites (36%) for the
percent of the site that was disturbed, and four sites (29%) for the percent of the
surrounding area that was disturbed.
Comparisons were made of the results obtained by different teams evaluating
seven of the pairs of created and natural wetlands. Whether the teams found similar
percent differences between the paired sites was determined. Again, differences
greater than 40% were noted for the three variables listed above. One of the 7 pairs
(14%) had differences greater than 40% for the percent of the site covered by open
water, four pairs (57%) for the percent of the site that was disturbed, and two pairs
(29%) for the percent of the surrounding area that was disturbed.
Finally, whether the mean difference between the created and natural sites was
greater than the mean difference between the teams was examined for those
wetlands sampled by only two teams. This was determined by taking the average of
the absolute value of the difference between the answers given for the 19 questions
on the site description form. The data were compared for sites where two teams
sampled the same site and for the pairs of sites. Forty percent of the site difference
averages were greater than the team difference averages, 1 5% were approximately
equal (within 1 %), and 45% were less. Therefore, for the sites which were sampled
by two teams, it appears that there was about as much difference observed between
created and natural sites as there was between teams.
Pielou
For three of the pairs of sites (evaluated by two teams each), the teams came
to the same conclusion for both test 1 and test 2. For one pair of sites (evaluated by
two teams), and for two pairs of sites (evaluated by four teams each), the teams
came to different conclusions for test 1. The teams did not always agree on whether
the sites had similar diversities.
23
-------
KOLL
315.500
1343.677
2371.854
3400.031
4428.208
+
+
C 2
C 1
C 3
C 4
N 3
N 1
N 4
N 2
NISQ
264.000
I
N 2
593.688
923.375
1253.063
1582.750
N 4
C 2
C 4
-t-
+
PACF
157.000
I
N 1
997.500
1
1838.000
2678.500
3519.000
N 3
C 1
C 3
Figure 2. Dendrograms produced from vegetation data collected at the three pairs of
freshwater wetlands. Dendrograms indicated that team differences were
less than wetland differences because the teams clustered before the sites.
C=Created wetland, N=Natural wetland; 1,2,3,4 are team numbers.
24
-------
LINC
111.000
I
1167.245
2223.490
3279.735
4335.979
4-
N 1
N 2
N 3
N 4
C 1
C 3
C 2
C 4
WILL
177.500
,
N 1
578.000
978.500
1379.000
1779.500
+
+
N 2
C 1
C 2
WEST
608.000
1151.438
1694.875
2238.313
2781.75
I-
+
N 1
N 2
C 1
C 2
Figure 3. Dendrograms produced from vegetation data collected at the three pairs
of salt marshes. Dendrograms indicated that team differences were less
than wetland differences because the teams clustered before the sites.
C=Created wetland; N=Natural wetland; 1,2,3,4 are team numbers.
25
-------
However, they consistently agreed that the created and natural sites were different.
The analyses determined that either the created and natural paired sites were not from
the same parent population or that the plant communities on the created sites were
not subsets of the communities on the natural sites.
WET
The total number of different ratings given by two teams at the same site was
compared with the total number of different ratings given to the paired created and
natural sites evaluated by the same team (Table 12). The total number of differences
in the evaluations was 20 out of a possible 240 for both of the comparisons.
Therefore, it appears that there was about as much difference in ratings between
teams as there was difference in ratings between pairs of sites for the sites where just
two teams visited.
26
-------
Table 12.
The number of differences in the Wetland Evaluation Technique (Adamus et
al. 1987) evaluations for pairs of created and natural wetlands and for teams
visiting the same wetland pair, C = created, N= natural.
SITE
TEAM
# OF DIFFERENCES
IN EVALUATIONS
SITE
C/N
# OF DIFFERENCES
IN EVALUATIONS
NISQ
2
0
NISQ
C
1
4
2
N
1
PACF
1
2
PACF
C
1
3
5
N
2
WEST
1
1
WEST
C
3
2
0
N
4
WILL
1
2
WILL
C
2
2
0
N
0
SEQU
3
2
SEQU
C
3
4
6
N
3
TOTAL
20
TOTAL
20
27
-------
SECTION 4
DISCUSSION
COMPARISON OF CREATED AND NATURAL WETLANDS
Freshwater wetlands
Different vegetation patterns were found at the three pairs of created and
natural freshwater wetlands. Possible reasons include the elimination of species found
only once within each site pair and the different number of teams sampling the sites.
Seventy-six percent of all species found during sampling were used in the vegetation
patterns analysis. By eliminating the less frequently found species, the results may
have been biased. In addition, the NISQ and PACF sites were sampled by two teams
and the KOLL site was sampled by four teams. Thus, the results were confounded
and exact comparisons were not possible. However, dendrograms from the cluster
analysis (Figure 2) showed that the teams consistently clustered before the created
and natural sites, indicating that the differences between the teams were not as great
as the differences between the sites.
Saltwater wetlands
The results indicated differences also existed between the three pairs of created
and natural saltwater sites. As with the freshwater sites, possible explanations
include the elimination of species found once within each site pair and different
numbers of teams sampling each site. The LINC site was sampled by all four teams
and the WEST and WILL sites were sampled by two teams each. The LINC site also
had a large proportion of unidentified plant species which may have affected the
results. Dendrograms for the pairs of saltwater sites (Figure 3) showed a similar
pattern as was found for the freshwater sites. The teams clustered before the created
and natural sites, indicating that the differences between the teams were not as great
as the differences between the sites.
COMPARISON OF TEAMS
In contrast to the above findings, the results from WET, the Pielou method, and
the site description data indicate that differences found between the teams were
about as prevalent as differences found between the created and natural wetlands.
The amount of heterogeneity on a site could have influenced the results. Although
criteria for transect placement were used, site heterogeneity would affect whether a
team sampled similar vegetation. Other possible reasons to explain this finding
28
-------
include; insufficient training, differing amounts of botanical expertise among the
teams, and subjective methods for data collection.
The teams would have benefitted from more complete field training before data
collection began. A one week training class in the use of WET, and the Army Corps
of Engineers' (COE) and EPA's wetland delineation methods was attended by all the
team members. However, time spent practicing the other field techniques and
ascertaining whether the teams understood the rationale for the methods and goals
of the study (i.e., why the data were collected in a certain order) was limited due to
the amount of information that had to be communicated during the course. A practice
run in the field using all the techniques would have helped both the team members
and instructors to identify potential problems and to promote open communication
early in the study.
The team members also had varying levels of botanical experience. The
different results found between the teams for the Pielou analysis could be partially
attributed to the varying skill levels of the botanists. The team with the most qualified
botanist provided the most detailed data sheets and plant specimens were carefully
collected and identified. This type of detailed documentation would likely be more
difficult for individuals with lesser botanical training.
Finally, all of the methods tested had a certain amount of inherent subjectivity
which could explain why teams sampling the same sites sometimes had different
responses. For example, the estimates given by different teams sampling the LINC
natural wetland for the percent of the wetland that was disturbed ranged between 0
and 100%. Although, the range for the estimates was smaller for the slightly less
subjective variable, percent of open water found on the wetland, it was still sizeable
(60%).
UTILITY OF METHODS
Pielou
The placement of the transects and determination of the number of species to
record, may have affected the results of the Pielou analysis. Each team determined
transect placement, therefore, the teams had different starting points and directions
for their transects. If the method is robust, then the transect placement should not
significantly effect results. Very heterogeneous sites, however, could produce varying
results. Also, the analysis calls for the determination of the number of species to
record in each quadrat, k, and assumes that there are always at least k different
species within two meters of each sampling point (Pielou 1986). However, the
analysis can be adversely affected by choosing a value for k that is either too small
or too large. If the number chosen is too small, then the method might underestimate
29
-------
the true diversity of the site. If the number chosen is too large, then difficulties
finding k species in the plot might be encountered. In this case, the method suggests
an alternative method for analysis which ultimately makes the quadrats more similar
to each other and the site diversity lower.
The Pielou method organized the data so that comparisons were made of the
most typical quadrat found at the sites. Although, this was done to make the
statistical analysis valid, information was lost by collapsing the 40 by 40 matrix into
one variable.
mi
WET did not distinguish between the saltwater sites for many of the functions
and it did not distinguish between either the fresh- or saltwater sites for four of the
twelve functions. Of course, there could have been few differences to distinguish
between. However, over 88% of the WET evaluations had low ratings for
groundwater recharge, nutrient removal/transformation, and wildlife d/a breeding, and
all the evaluations were medium for production export. For the remaining functions,
the evaluations were different for various reasons, e.g., different wetland types,
different paired created and natural sites, and different teams at the same site.
WET was evaluated by the teams at the end of the study and during a quality
assurance audit of the teams in the field. The evaluation of WET by the field crews
was important to the study because the field crews could be considered typical users
of the technique, i.e., they were members of government agencies who had some
training in the technique. However, the following discussion is based solely on their
opinions. No doubt other users would have different experiences and opinions about
WET.
The teams felt that many of the questions were too complicated and time
consuming, and that some of the questions were difficult to interpret or did not make
sense (D. Coffey, ManTech Environmental Technology Inc., Corvallis, OR, pers.
comm.). They also felt that there should have been an option which enabled them to
avoid answering questions or evaluating functions that were not applicable to a given
region or situation (K. Kunz, COE, Seattle, WA, pers. comm.).
The consensus of the teams was that the WET data were not sensitive enough
to adequately assess the success, or lack thereof, of the creation projects. They felt
that the probability of a wetland performing a function was not the same as
determining if a created wetland was actually functioning (K. Kunz, COE, Seattle, WA,
pers. comm.).
The teams felt that WET was not ready to be adopted as a national procedure.
They questioned whether the interpretation of the literature which was used to
30
-------
develop the predictors could be extrapolated to the Pacific Northwest and whether
generalized wetland questions could adequately characterize functions and values in
extremely different wetland types (K. Kunz, COE, Seattle, WA, pers. comm.).
In a report prepared for the Federal Highway Administration, Normandeau
Associates, Incorporated (NAI) provided a review and critique of WET (NAI 1990).
NAI also found that certain questions were worded in such a way as to make them
easy to misinterpret and that WET was difficult to use due to the obtuse wording of
the predictor questions. They suggested that WET was too general for mitigation
evaluation, primarily because it placed more importance on location than wetland
features. NAI also found that for some of the functions (sediment/toxicant retention,
nutrient removal/transformation, aquatic d/a and wildlife d/a breeding), the "recent
alteration" factor was pivotal in determining the probability rating for a wetland and
almost always lead to a low probability rating. The low rating did not reflect the stage
of development or design features, but rather was strictly a function of time (NAI
1990). It does not appear that the "recent alteration" factor affected the data in this
report because there were some high, medium and low ratings for the above listed
functions for both the created and natural wetlands. Finally, NAI, the team members,
and others (e.g., Odum et al. 1986, van der Valk 1989) have all expressed concerns
about the use of WET in different regions.
RECOMMENDATIONS
The lessons learned from this study can be summarized into the following
recommendations:
1. Reserve an adequate amount of time for training to increase the probability that
comparable results are obtained by the teams. Strive to reduce the variability
between teams by allowing them the opportunity to practice, to discuss any
areas of difficulty or disagreement, and to practice again. Having one team do
the sampling would eliminate the between team variability, but it would also
eliminate any determination of the method's replicability.
2. Evaluate between team performance before beginning actual sampling.
3. Ensure that the directions for all procedures are clear and easy to follow.
4. Strive to reduce the varying levels of experience and expertise on the teams.
Realize that if there are varying levels of experience and expertise, variable
results might be encountered.
5. Make sure communication between team members and researchers is initiated
early so that potential problems are discovered.
31
-------
6. Ensure that problems with the procedures are documented in writing so that
they can be corrected for future studies.
CONCLUSIONS
The goals of the study were difficult to evaluate due to differences in the
results of teams visiting the same wetland. The heterogeneity within the sites could
explain some of the differences. If the sites were heterogeneous, then team
differences might be expected. However, it is difficult to determine whether
differences found between teams at the same site were due to site heterogeneity or
actual differences between the teams.
The performance of WET was also difficult to assess due to different team
results at the same site. However, concerns were expressed about WET by the teams
participating in this study and NAI (1990). Both felt that WET: included questions
that were complex and difficult to interpret, did not address regional concerns, and
was not sensitive enough to assess success of created wetlands. Therefore, the
applicability of WET for comparing created and natural wetlands is not clear at this
point.
Team differences also made determination of the applicability of the Pielou
method (1986) difficult. Concerns about the Pielou method included choosing the
value of k (the number of species to sample) and the potential loss of information
during analysis.
Suggested improvements for sample design would be to have the same number
of teams visit each site. This makes comparison among teams more straightforward.
Also, comparing samples of wetlands is more meaningful than comparing paired
wetlands because extrapolation to the population of interest is possible. However,
use of the Pielou method requires paired sites, so one of the goals of this study, to
test techniques, could not have been met with another sampling scheme.
The objectives of this study were to test the effectiveness of methods that
could be used by agencies to evaluate created wetlands, to use these methods to
compare the created and natural wetlands, and to evaluate the consistency of results
observed by teams sampling the same site. The limitations of both WET and the
method recommended by Pielou (1986) were identified and documented in a "real life"
situation. Because agency personnel, given the constraints of their work situation,
conducted the study, we feel that the results of this study will have some utility for
regulators in wetland evaluation. Although what we can say about the similarities and
differences between the created and natural sites is restricted, the information
compiled on the sites should provide a baseline for future evaluations.
32
-------
SECTION 5
LITERATURE CITED
Adamus, P.R. 1988. Criteria for created or restored wetlands, p. 369-372. ]n: Hook,
D.D., W.H. Mckee, Jr., H.K. Smith, J. Gregory, V.G. Burreii, Jr., M.R. Devoe,
R.E. Sojka, S. Gilbert, R. Banks, L.H. Stolzy, C. Brooks, T.D. Mathews, and
T.H. Shear. (Eds.). The Ecology and Management of Wetlands. Vol. 2:
Management, Use and Value of Wetlands. Croom Helm, London and Sydney.
Adamus, P.R., E.J. Clairain, Jr., R.D. Smith, and R.E. Young. 1987. Wetland
Evaluation Technique (WET), Vol. II: Methodology. Operation Draft TRY-87_,
U.S. Army Corps of Engineers, Waterways Experiment Station, Mississippi.
Hitchcock, C.L. and A. Cronquist. 1981. Flora of the Pacific Northwest--An
Illustrated Manual. 5th Edition. University of WA Press, Seattle, WA.
Kentula, M.E., J.C. Sifneos, J.W. Good, M. Rylko, and K. Kunz. In press. Trends and
patterns in Section 404 permitting requiring compensatory mitigation in Oregon
and Washington. Env. Manag.
Kusler, J.A. and M.E. Kentula. 1990. Executive Summary, p. xvii-xxv. In J.A.
Kusler and M.E. Kentula (Eds.), Wetland Creation and Restoration: The Status
of the Science. Island Press, Washington, D.C.
Normandeau Associates, Inc. 1990. A user's critique of the Wetland Evaluation
Technique. Draft prepared for the Federal Highway Administration,
Washington, D.C.
Odum, W.E., J. Harvey, L. Rozas, and R. Chambers. 1986. The functional
assessment of selected wetlands of Chincoteague Island, Virginia. U.S. Fish
Wild. Serv. NWRC Open File Rep. 86-87.
Pielou, E.C. 1986. Assessing the diversity and composition of restored vegetation.
Can. J. Bot. 64:1344-1348.
Quammen, M.L. 1986. Measuring the success of wetlands mitigation. Nat.
Wetlands Newsl. 8(5):6-8.
Reed, P.B., Jr. 1988. National List of Plant Species that Occur in Wetlands:
Northwest (Region 9). U.S. Fish Wildl. Serv. Biol. Rep. 88(26.9). Washington,
D.C.
van der Valk, A.G. (Ed.). 1989. Northern Prairie Wetlands. Iowa State University
Press, Ames, Iowa.
33
-------
APPENDIX I
DESCRIPTIONS OF CREATED AND NATURAL WETLANDS
USED IN THE WASHINGTON FIELD STUDY
The following information on the created wetlands was compiled from the Section
404 permit record (Kentula et. al submitted). The natural wetlands were selected and
described by Michael Rylko and Kathy Kunz. Four letter codes (e.g., KOLL) are
pseudonyms for the permit numbers and are used to preserve the anonymity of the
sites.
FRESHWATER WETLANDS
KOLL
wetland type
location
area (acres)
permit issued
age when sampled
palustrine emergent marsh
North Creek (Sammamish River)
5.3
1984
9 months
The natural wetland was located 3 miles north west of the created wetland in a
similar situation.
NISQ
wetland type
location
area (acres)
permit issued
age when sampled
palustrine emergent marsh
Nisqually River, side channel
0.1
1985
7 months
The natural wetland was located to the immediate north of the created site in a similar
habitat.
34
-------
PACF
wetland type
location
area (acres)
permit issued
age when sampled
paiustrine emergent marsh
Woodward Creek
0.1
1985
9 months
The natural wetland was located to the immediate north of the created site in a similar
habitat.
35
-------
SALTWATER WETLANDS
LINC
wetland type
location
area (acres)
permit issued
age when sampled
salt marsh
Puyallup River
3.8
1984
13 months
The created wetland was located in a heavily urbanized section of the Puyallup River
and there were no comparable natural sites in the immediate vicinity. The only natural
wetland in a similar situation was located near Seattle in the Duwamish River Basin
at Kellogg Island. Although the areas were some 30 miles apart, both were located
in a heavily urbanized tidal river with similar species composition.
WEST
wetland type
location
area (acres)
permit issued
age when sampled
salt marsh
Gray's Harbor
0.1
1986
8 months
The natural wetland was located to the immediate north, east, and south of the
created wetland in similar habitat.
WILL
wetland type
location
area (acres)
permit issued
age when sampled
salt marsh
Willapa Bay
2.2
1983
22 months
The natural wetland was located to the immediate north of the created site in a similar
habitat.
36
-------
SEQU
wetland type
location
area (acres)
permit issued
age when sampled
eelgrass beds
Sequim Bay
2.0
1983
27 months
The natural wetland was 1 mile north of the created wetland within the confines of
Sequim Bay.
BLR1
wetland type
location
area (acres)
permit issued
age when sampled
mudflat
Blair Waterway (Commencement Bay)
0.3
1984
28 months
The natural wetland was located on the Hylebos Waterway in a very similar situation
to the created wetland.
PIER
wetland type
location
area (acres)
permit issued
age when sampled
mudflat
Shilshole Bay
0.5
1986
8 months
The natural wetland was located offsite at Alki Point, West Seattle, approximately 10
miles from the created wetland.
37
-------
APPENDIX II
SPECIES FOUND DURING THE WASHINGTON FIELD STUDY
The following is a list of plant species found in the Washington Field
Study. The species are listed in phylogenetic order. Hitchcock and
Cronquist (1981) was used to identify species found during sampling-
Adjacent to the specie's names are the codes used in the analyses.
The plant's indicator status and native/introduced determination were
taken from The Regional List of Plant Species that Occur in Wetlands:
Northwest (Region 9) (Reed, 1988). Codes are: "o"—obligate wetland
species; "w"--facultative wetland species; "f"--facultative wetland
species; "u"—facultative upland species; "p"—upland species; "a"—
absent from the List of Species that Occur in Wetlands; "+"—upper end
of category; —lower end of category; "\M—intermediate within the
category; "n"—native species; "i" — introduced species; ***-—no
information.
Code Vegetation Found
010000
Equisetaceae
010100
* * *
Equisetum sp.
010101
w\n
Equisetum arvense
010103
w\n
Equisetum telmateia
020000
Polypodiaceae
020101
f\n
Athyrium filix-femina
020200
* * *
Dryopteris sp.
020301
P\n
Polystichum munitum
020401
u\n
Pteridium aquilinum
040000
Cupressaceae
040101
f \n
Thuja plicata
050000
Pinaceae
050101
f\n
Picea sitchensis
050201
p\n
Pseudotsuga menziesii
050301
u-n
Tsuga heterophylla
060000
Salicaceae
060101
w\n
Populus balsamifera
060102
f+i
Populus tremuloides\Populus tremula tremuloides
060102
f+i
Populus tremuloides
060102
f+i
Populus tremula tremuloides
060103
w\n
Populus trichocarpa\Populus balsamifera trichocarpa
060103
w\n
Populus trichocarpa
060103
w\n
Populus balsamifera trichocarpa
060200
~ * *
Salix sp.
060202
w+n
Salix geyeriana
060203
w-n
Salix hookeriana
060204
w+n
Salix lasiandra
060205
w\n
Salix piperi
060207
f\n
Salix scouleriana
060208
w\n
Salix sessilifolia
060210
w\n
Salix sitchensis
. 38
-------
070000
Betulaceae
070101
f\n
Alnus rubra
080000
Urticaceae
080101
f + i
Urtica dioica
090000
Polygonaceae
090100
***
Polygonum sp.
090101
w-i
Polygonum aviculare
090102
o\n
Polygonum coccineum\Polygonum amphibiura emersuir,
090102
o\n
Polygonum coccineum
090102
o\n
Polygonum amphibium emersum
090104
o\i
Polygonum hydropiper
090106.
w+n
Polygonum lapathifolium
090107
wdi
Polygonum persicaria
090200
***
Runex sp.
090201
u\i
Rumex acetosella
090202
w\i
Rumex conglomeratus
090203
w\i
Rumex crispus
090204
w+n
Rumex occidentalis
090205
w+n
Rumex maritimus
100000
Chenopodiaceae
100101
w\n
Atriplex patula
100204
pU
Chenopodium foliosum
100205
w+n
Chenopodium rubrura
100301
o\n
Salicornia virginica
120000
Garyophyllaceae
120201
w\n
Spergularia canadensis
120202
odn
Spergularia marina
120301
w+n
Stellaria calycantha
120303
o\n
Stellaria humifusa
150000
Ranunculaceae
150102
w\i
Ranunculus repens
160000
Brassicaceae
160601
w+n
Rorippa curvisiliqua
160603
o\i
Rorippa nasturtium-aquaticura\Nasturtium officinale
160603
o\i
Rorippa nasturtium-aquaticum
160603
o\ 1
Nasturtium officinale
180000
Saxifragaceae
180100
***
Mitella sp.
180301
f-n
Tiarella trifoliata
180401
f\n
Tolmiea menziesii
190000
Grossulariaceae
190101
f+n
Ribes lacustre
200000
Rosaceae
200101
P\i
Crataegus monogyna
200301
w+n
Geum macrophyllura
200401
u\n
Oemleria cerasiformis
200501
f+n
Physocarpus capitatus
200601
o\n
Potentilla pacifica\Potentilia anserina
200601
o\n
Potentilla pacifica
200601
o\n
Potentilla anserina
200700
* * A
Prunus sp.
200801
pU
Pyrus malus\Malus sylvestris
39
-------
200801 p\i Pyrus malus
200801 p\i Malus sylvestris
200903 u\n Rosa rugosa
201001 u-i Rubus discolor
201002 u+i Rubus laciniatus
201003 f\n Rubus pubescens
201004 f\n Rubus spectabilis
201005 u\n Rubus ursinus
201101 p\i Sorbus aucuparia
201201 w\n Spiraea douglasii
210000 Fabaceae
210101 u\i Cytisus scoparius
210301 f\i Lotus corniculatus
210302 u\n Lotus micranthus
210500 *** Trifolium sp.
210502 p\i Trifolium dubium
210505 u+i Trifolium repens
210507 w+n Trifolium wormskjoldii
210603 u\n Vicia gigantea
210604 p\i Vicia hirsuta
210605 p\i Vicia sativa
250000 Callitrichaceae
250100 *** Callitriche sp.
250102 o\n Callitriche verna
260000 Aceraceae
260101 u+n Acer circinatum
260102 u\n Acer macrophyllum
280000 Rhamnaceae
280101 p\i Rhamnus purshiana
300000 Hypericaceae
300101 o\n Hypericum anagalloides
300103 p\i Hypericum perforatum
320000 Onagraceae
320201 w\n Circaea alpina
320300 *** Epilobium sp.
320301 u+n Epilobium angustifoliura
320304 w-n Epilobium watsonii\Epilobium
320304 w-n Epilobium watsonii
320304 w-n Epilobium ciliatum watsonii
340000 Araliaceae
340101 p\i Hedera helix
350000 Apiaceae
350101 f\n Angelica lucida
350301 o\n Cicuta douglasii
350501 o\n Lilaeopsis occidentalis
350601 o\i Oenanthe sarmentosa
360000 Cornaceae
360101 w\n Cornus stolonifera
370000 Primulaceae
370201 w+i Glaux maritima
390000 Gentianaceae
390100 *** Centaurium sp.
ciliatum watsonii
40
-------
390101
w\n
Centaurium muhlenbergii
410000
Cuscutaceae\Convolvulaceae
410000
Cuscutacea sp.
410000
Convolvulaceae sp.
410101
w\n
Cuscuta salina
430000
Boraginaceae
430102
o\n
Myosotis laxa
440000
Lamiaceae
440101
a\i
Galeopsis tetrahit
440201
u+i
Glecoma hederacea
440801
w\n
Stachys cooleyae\Stachys e
440801
w\n
Stachys cooleyae
440801
w\n
Stachys emersonii
450000
Solanaceae
450101
u\i
Atropa belladonna
450201
f\i
Solanum dulcamara
460000
Scrophulariaceae
460702
w+n
Mimulus rooschatus
460801
w+n
Orthocarpus castillejoides
460901
f-i
Parentucellia viscosa
461101
***
Veronica sp.
461101
o\n
Veronica americana
461102
odn
Veronica peregrina
461104
o\i
Veronica scutellata
470000
Plantaginaceae
470101
w\ i
Plantago coronopus
470102
u+i
Plantago lanceolata
470103
f+i
Plantago major
470104
w+n
Plantago maritima
480000
Rubiaceae
480100
***
Galivun sp.
480101
u\n
Galium aparine
480102
w+n
Galium trifidum
490000
Caprifoliaceae
490101
f\n
Lonicera involucrata
490201
u\n
Sambucus racemosa
490301
u\n
Symphoricarpos albus
510000
Campanulaceae
510101
p\n
Triodanis perfoliata
520000
Asteraceae
520101
u\n
Achillea millefolium
520201
u\i
Anthemis cotula
520401
w\n
Aster subspicatus
520501
w\n
Bidens cernua
520900
***
Cirsium sp.
520901
u+i
Cirsium arvense
520902
w-n
Cirsium edule
520903
u\i
Cirsium vulgare
521101
w+i
Cotula coronopifolia
521201
P\i
Crepis capillaris
521301
f+i
Gnaphalium chilense
521302
f+n
Gnaphalium palustre
41
-------
52
52
52
52
52
52
52
52
52
52
52
52
52
52
52
53
53
53
54
54
55
55
55
57
57
58
58
58
58
58
58
58
58
58
58
58
58
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
1
a\n Gnaphalium microcephalum
f+n Gnaphalium uliginosum\Filaginella uliginosa
f+n Gnaphalium uliginosum
f+n Filaginella uliginosa
w\n Grindelia integrifolia
p\i Hypochaeris glabra
p\i Hypochaeris radicata
f-n Lactuca serriola
p\i Senecio jacobaea
*** Solidago sp.
u\n Solidago canadensis
f-i Sonchus asper
p\i Sonchus oleraceus
o\n Jaumea carnosa
u\i Taraxacum officinale
Alismataceae
o\n Alisma plantago-aquatica
o\n Sagittaria latifolia
Hydrocharitaceae
*** Elodea sp.
Juncaginaceae
o\n Triglochin concinnum
o\n Triglochin maritimum
Zosteraceae
o\i Zostera nana
Juncaceae
*** Juncaceae 1
*** Juncaceae 2
*** Juncus sp.
o\n Juncus articulatus
o\n Juncus balticus
w+n Juncus bufonius
w+n Juncus effusus
w\n Juncus ensifolius
w\n Juncus nevadensis
w+n Juncus oxymeris
f\n Juncus tenuis
Cyperaceae
*** Cyperaceae 1
*** Carex sp.
*** Carex sp. 1
*** Carex sp. 2
*** Carex sp. 3
*** Carex sp. 4
f+n Carex deweyana
o\n Carex lyngbyei
o\n Carex obnupta
o\n Carex stipata
*** Eleocharis sp.
w\n Eleocharis bolanderi
o\n Eleocharis ovata
o\n Eleocharis palustris
42
-------
590306
o\n
Eleocharis parvula
590400
***
Scirpus sp.
590401
o\n
Scirpus acutus
590402
o\n
Scirpus americanus
590404
o\n
Scirpus maritimus
590405
o\n
Scirpus microcarpus
500406
o\n
Scirpus validus
600000
Poaceae
60UN01
***
Poaceae 1
60UN02
***
Poaceae 2
60UN03
***
Poaceae 3
600102
u\i
Agropyron repens
600200
~ ~~
Agrostis sp.
600201
w\i
Agrostis alba
600202
w\n
Agrostis exarata
600204
f\n
Agrostis scabra
600205
P\i
Agrostis tenuis
600402
w\i
Alopecurus pratensis
600501
u\i
Ammophila arenaria
600601
u\i
Anthoxanthum odoratum
600903
P\i
Broitius rigidus
601001
u\i
Dactylis glomerata
601201
w\n
Deschampsia cespitosa
601401
w\n
Distichlis spicata
601601
p\n
Elymus cinereus
601801
u-i
Festuca arundinacea
601803
fdi
Festuca myuros\Vulpia myuros
601803
fdi
Festuca myuros
601803
fdi
Vulpia myuros
601804
P\n
Festuca ovina
601805
f \n
Festuca rubra
601902
w+n
Glyceria elata
601903
°\i
Glyceria grandis\Glyceria maxima grandis
601903
°\i
Glyceria grandis
601903
o\i
Glyceria maxima grandis
601904
o\n
Glyceria leptostachya
602101
f\n
Holcus lanatus
602200
***
Hordeum sp.
602201
w\n
Hordeum brachyantherum
602401
u\i
Lolium multiflorum\Lolium perenne multiflorum
602401
u\i
Lolium multiflorum
602401
u\i
Lolium perenne multiflorum
602701
w\n
Phalaris arundinacea
602801
u\i
Phleum pratense
602900
***
Phragmites communis
603003
f\n
Poa palustris
603004
u+n
Poa pratensis
603005
w-i
Poa trivialis
620000
Typhaceae
620101
o\n
Typha latifolia
630000
Araceae
630101
o\n
Lysichitum americanum\Lysichiton americanus
43
-------
630101
o\n
Lysichituin americanum
630101
o\n
Lysichiton aroericanus
664000
Lemnaceae
640101
o\n
Lemna minor
650000
Liliaceae
650301
f-n
Maianthemum dilatatum
650401
f-n
Smilacina stellata
660000
Iridaceae
660100
aa a
Iris sp.
660101
°\i
Iris pseudocorus
67UN00
a a a
Unknowns
67UN01
aaa
Unknown
1
6 7 UN 0 2
aaa
Unknown
2
67UN03
aaa
Unknown
3
67UN04
AAA
Unknown
4
67UN05
AAA
Unknown
5
67UN06
AAA
Unknown
6
68WC01
AAA
Unknown
7
68WC02
AAA
Unknown
8
-------
TECHNICAL REPORT DATA
(Piatt nnf Jmttrycnoru on the rtvmt or [art comj*
>
!. REPORT NO. >.
EPA/600/R-92/013
4. T|Tl| ANQ SUBTITLE
A Pilot Study to Compare Created and
Natural Wetlands in Western Washington
and Evaluate Methods
1 REPORT DATE
January 1992
B. PERFORMING ORGANIZATION CODE
J. AUTnORISI
Jean C. Sifneos, Donna L. Frostholm,
and Mary E. Kentula
I. PERFORMINO ORGANIZATION REPORT NO
r PERFORMING ORGANIZATION NAM! AND ADORESS
1st two, ManTech Technology Services,
ERL-Corvallis, OR; Kentula, US EPA, ERL-
Corvallis.
IB. PROORAM ELEMENT NO.
VI. B6klf hACT/ORANT NO.
12. SPONSORING AGENCT NAME AND ADOREBS
US Environmental Protection Agency
Environmental Research Laboratory
200 SW 35th Street
Corvallis, OR 97333
11. TYPE OF REPORT AND PERIOD COVERED
Published Report
14. SPONSORING AGENCY CODE
EPA/600/02
1991. U.S. Environmental Protection Agency, Environmental Research
{^Laboratory, Corvallis, OR
IS. abstract
Nine created wetlands were paired with nine natural wetlands and compared for
species composition, species diversity, wetland function and other site
characteristics. Results from comparison of species composition and species diversity
indicated that some differences existed between the created and natural sites.
However, because differences also existed in the data collected from different teams
sampling the same site,, the created and natural site differences were confounded.
In addition to the possible heterogeneity within the sites, low repl icability between
the teams may have been due to insufficient training prior to field work, different
botanical skill levels, or the subjectivity of some of other data collection forms.
comparison of species composition also found some similarities
with respect to the native/introduced and indicator status of the species found at
the created and natural sites..v^\Generally, the species classified as obligate,
facultative wetlands, and wetland were native to the Pacific Northwest, while those
species classified as facultative up\and and upland were introduced.
l7">
IT. KEV WORDS AND DOCUMENT ANALYSIS
1. DESCRIPTORS
fc.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT ( FcU/Croup
created wetlands, Clean Water Act
Section 404, mitigation, Washingt
>n,
It DISTRIBUTION STATEMENT
Release to Public
IB. SIGpRiTir Ci,aSS (ThitMtportj
Unciassifiea
>1. NO. or PAGES
52 pqs
M SECURITY CLASS (Thiiptt!
Unclassified
aa. price
OA F*rm lit0-1 (fr-91)
------- |