United States Office of Research and EPA/600/R-92/150
Environmental Protection Development August 1992
Agency Washington, DC 20460
-------
-------
op
6
^S^t
CO <
CO
o
Q_
DC
L1J
CL
g
'ffl
CD CO
O) &
< cr
S "eo
>
CD OD
E «o
CD
CO
'
CD
W tS
S >
O
LO
ca
CO
•»
3 O
m -^
g H5 c -jg^
•s > 1 '8 i"g
d C uJ .— l£z m pn
13 UJ O O OQ-S
E
^
o
I
CL
LU
-------
-------
EPA/600/R-92/150
; August 1992
AN APPROACH TO IMPROVING
DECISION MAKING IN WETLAND RESTORATION
AND CREATION
Authors:
Mary E. Kentula1
Robert P. Brooks2
Stephanie E. Gwin3 [jf
Cindy C. Holland3 . .
Arthur D. Sherman3 ^
Jean C. Sifneos3 . ^
Editor: g
Ann J. Hairston3 £3
•••.'• o
1 U.S. Environmental Protection Agency ; uj
Environmental Research Laboratory ffl
200 SW 35th Street • : gj
Corvallis, OR 97333 ; °^
2The Pennsylvania State University
Forest Resources Laboratory
University Park, PA 16802
3ManTech Environmental Technology, Inc.
USEPA Environmental Research Laboratory
. 200 SW 35th Street
Corvallis, OR 97333
Project Officer • Mary E. Kentula -Wetlands Research Program
U.S. Environmental Protection Agency • Environmental Research Laboratory
200 S.W. 35th Street - Corvallis, OR 97333
Printed on Recycled Paper
-------
As wetland losses continue and restoration and creation efforts increase,
Ibn&tertn research data become essential to understanding the impacts of our
regulatory decisions. AN APPROACH TO IMPROVING DECISION MAKING
IN WETLAND RESTORATION AND CREATION is the culmination of five
years of research primarily in Connecticut, Florida, and Oregon. This research
compares populations of natural and created wetlands to determine whether
restored wetlands successfully replace wetlands lost to development and other
pressuresT-JClaetype of information synthesized in this document can be used
by resource managers in determining strategies for mitigation of wetland loss-
es In addition, this approach addresses management concerns such as site se-
lection for future restoration projects, assessment of the level of attainable
function 'for restored wetlands, and how to evaluate when the desired level of
function has been achieved.
Although primarily designed to meet the needs of the EPA regions and the
Office of Water, this is a useful document that will undoubtedly be read with
varying expectations by a wide audience. There are ideas for all readers. The
approach offered makes a significant contribution to the scientific information
base for decisions on wetland restoration and creation. This is a synthesis
document and much additional information can be found by consulting the
original research results. This approach is not intended to define EPA policy.
And, although we do not endorse any one approach, we can certainly endorse
the main theme of this document, "we must learn from what we have done
and use that information to improve future resource management".
Wetlands Division
Office of Wetlands, Oceans, and Watersheds
An Approach to Improving Decision Making in Wetland Restoration and Creation
\
-------
DISCLAIMER
The research described in this document has been, funded by the United
States Environmental Protection Agency under Contract *68-C8-0006 to Man-
Tech Environmental Technology, Inc. and Contract #68-CO-0021 to Technical
Resources, Inc. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
This document should be cited as:
Kentula, M.E., R.P. Brooks, S.E. Gwin, C.C. Holland, A.D. Sherman, and
J.C Sifneos. 1992. An Approach to Improving Decision Making in Wetland
Restoration and Creation. Edited by A.J. Hairston., U.S. Environmental Protec-
tion Agency, Environmental Research Laboratory, Corvallis, OR.
-------
LETTER
To the Reader,
We have kept you in mind throughout the preparation of DECISION MAK-
ING In particular, we have attempted to fill a gap in the information available
on wetland restoration and creation by addressing how to improve future deci-
sions by evaluating the results of past decisions. We are aware of the varied
needs and interests of our audience and feel a responsibility to meet the ex-
pectations those needs and interests bring to a reading of our work. Reflecting
this diversity, the reviews of our draft manuscript indicated that many readers
had a "favorite" or "most useful" chapter and that there was little agreement as
to which of the chapters it was! We learned that DECISION MAKING could
be read and used in a number of ways we had not anticipated.
We recommend that you review the Table of Contents before beginning to
read the book and use the chapter and section titles to select the parts that will
be of most interest to you. Each chapter was written both to fit into the frame-
work of the book and to present a single concept. DECISION MAKING does
not have to be read in order, from front to back. The only exception is Chap-
ter 1, which should be read first because it defines the terms we use. After
that, if you are interested in monitoring projects, go to Chapter 4, OR jump to
Chapter 6 to read about evaluating design, OR go on to Chapter 2. Whichever
route you choose, we hope that you will find much of value.
The Authors
An Approach to Improving Decision Making in Wetland Restoration and Creation
iii
-------
EXECUTIVE SUMMARY
AN APPROACH TO IMPROVING DECISION MAKING
IN WETLAND RESTORATION AND CREATION
: The U.S. Environmental Protection Agency's Wetlands Research Program
(WRP) has developed an approach to improving decision making in wetland
restoration and creation projects. The WRP Approach uses data from a moni-
tbring program, including both natural wetlands and those restored and creat-
ed; tb develop performance criteria, track the development of projects, and
suggest improvements in the design of future projects. For the past five years,
scientists in association with the WRP have been developing the Approach by
comparing the characteristics of mitigation projects and natural wetlands to
test methods for data collection, and to evaluate project design and compli-
arice: with permit conditions. Many of the same methods were used in studies
iri Connecticut; Tamp'a, Florida; Portland, Oregon; and Seaside, Oregon, so
that the techniques could be evaluated/the findings from all studies compared
arid'the results used to refine the Approach.
' The projects studied were typically less than or equal to five years old, and
the majority were what is probably the most common freshwater mitigation
project nationally—a pond with a fringe of.emergent marsh. We chose this
type of project because they were abundant, comprising a major proportion of
the compensatory mitigation projects required nationally under Section 404 of
the Clean Water Act. Because the Approach was developed in freshwater sys-
tems, the monitoring techniques and examples presented will transfer most
readily to freshwater nontidal wetlands. However, application of the Ap-
proach is not limited to either mitigation projects or freshwater nontidal wet-
lands; it is determined by the needs of the agency or organization involved,
and ultimately, the status of the wetland resource.
The Approach is based on the assumption that natural wetlands in a region
can be used as models to define the standards for restoration and creation pro-
jects. Comparison of wetland projects with natural wetlands located in a simi-
lar land use setting and, therefore exposed to similar ecological conditions, is
important to ensure that what is "expected" of a project is within the bounds of
possible performance. Major recommendations are:
Use information in project files to guide decision making.
Target areas at greatest risk.
Base the level of effort used in monitoring on information needs.
An Approach to Improving Decision Making in Wetland Restoration and Creation
V
-------
EXECUTIVE SUMMARY
i
• Consider the landscape setting of the wetlands when defining the
• populations to be compared.
• Use the characteristics of natural wetlands and wetland projects to
define the standard. !
- Make the process of setting performance criteria and defining design
guidelines iterative.
.CHAPTER 1 presents an overview of the WRP Approach. This includes
discussion of the above recommendations and the major analytical tool of the
Approach, the performance curve. The performance curve documents the de-
velopment of the ecological function of wetland projects over time relative to
similar natural wetlands. We envision that a set of performance curves will be
produced for each function or indicator measured. What is measured is deter-
mined by the goals of the resource management program and the specific pro-
jects. Management questions that can. be answered using this strategy include:
• What level of function is achievable for natural wetlands and wetland
projects in a particular land use setting?
- Do the projects achieve the level of function of natural wetlands in
similar settings?.
- . How long does it take for projects to achieve the desired level of
function?
In a mitigation context, answers to such questions would allow managers
to identify which permits should 1) be most critically reviewed because of low
probability of successful mitigation; 2) require the most comprehensive checks
on design and implementation of the mitigation project because of uncertain
probability of success; and 3) require minimal checks on design and imple-
mentation of the mitigation project because of high probability of success.
CHAPTER 2 details how to use the information from project files in deci-
sion making. If the data are to be of use in protecting the wetland resource,
they must be updated, compiled, analyzed, and reported. For example, analy-
sis of previous trends in permitting can reveal locales and wetland types sub-
ject to the most intense permitting activity. With knowledge of such trends,
permitting agencies can take action to avoid potential losses in wetland num-
ber, type, function, and area.
VI
-------
EXECUTIVE SUMMARY
CHAPTER 3 describes a method for sampling populations of projects and
natural wetlands to select sites for study. Targeting efforts to.areas,where the
resource is/ or is predicted to be, at risk is also discussed. The Approach de-
scribes how to 1) define the population of projects to be sariripled; 2) use the
location of the sites to define the boundaries of the study area within the area
at risk; 3) use the characteristics of the population of projects to define the
population of natural wetlands to be sampled; 4) randomly 'select sites from
the populations of projects and natural wetlands; and 5) finalize the list of sites
to be sampled by verifying that the sites exist, are accessible, :and belong to the
populations defined. :
CHAPTER 4 presents a post-construction monitoring strategy that recom-
mends three levels of sampling depending on the age and goals of the project:
documentation of as-built conditions; routine assessments; and comprehensive
assessments. For each of the variables suggested, a brief rationale relating it to
wetland function is provided. Finally, components integral to a post-construc-
tion monitoring plan, such as maintaining data quality, timing of sample col-
lection, and controlling damage to the site, are discussed.
A major obstacle to long-term monitoring and, therefore, implementation
of the WRP Approach is cost. The special insert, Volunteers and Natural Re-
source Monitoring, that follows Chapter 4, presents a possible low-cost, high-
profit solution. WRP scientists worked closely with Neal Maine, an award
winning science education specialist, to train and coordinate a group of citi-
zen volunteers to assist in the study of the Trail's End mitigation project near
Seaside, Oregon. The study makes the connection between scientific'research
and public education which is so often ignored. The research profited be-
cause more types of data were collected over a longer period of time than
would have been possible otherwise. The community profited because volun-
teers who were teachers taught their students to collect data on local wetlands
using the techniques learned from the study. •"• ; ; "
In CHAPTER 5 four different types of graphs are suggested for representing
monitoring data: performance curves, summary or descriptive graphs, time se-
ries graphs, and characterization curves.' Statistical methods are outlined for
data manipulation that will enable resource managers to organize incoming
data to track the progress of projects, and to develop criteria for the evaluation
of future projects. Graphical displays are used to'illustrate how to evaluate
projects and set performance criteria.
CHAPTER 6 illustrates how data collected from local natural Wetlands can,
and should/be used to improve the design of projects. This chapter details de-
sign features, including type of wetland; slopes of banks; amount of area; and
An Approach to Improving Decision Making in Wetland Restoration and Creation
-------
EXECUTIVE SUMMARY
appropriate hydrology, vegetation, and soils/substrates. Development of a
planting list of species appropriate to a specific wetland type and locale is pre-
sented as an example of how to tailor project design to meet local needs.
We have developed the WRP Approach to help anyone working to protect
the wetland resource use past data from restoration or creation projects as a
management tool to improve decision making and, thereby, the ability to re-
store and create wetlands in the future. Our philosophy is that by considering
the surrounding land use, comparable natural wetlands, and similar projects,
you can design wetland projects with a better chance of long-term success.
Determining the effects of different land uses on wetland function will be a
major theme of our upcoming research. Such information will be important to
both the protection of the wetland resource and the success of restoration and
creation projects. With knowledge of the effects of surrounding land uses, ap-
propriate management strategies can be employed to protect key wetlands In
addition, knowing how present and projected development of an area will af-
fect wetland function can influence decisions on how to prioritize restoration
sites for maximum ecological benefits.
Fundamentally, as we plan and implement!new studies we will continue
to treat existing projects as experiments in progress and to promote the idea
that we all must "...learn by going where we need to go..." (Roethke 1961).
Vlll
-------
CONTENTS
PREFACE..... ....,.., • '
DISCLAIMER , • •'"
TO THE READER..... • r— "i
EXECUTIVE SUMMARY , v
LIST OF TABLES :».«•. *ii
LIST OF FIGURES. ......! •••• •-• xiii
BACKGROUND AND ACKNOWLEDGEMENTS .„ .....'...'..... .....xvii
. AN UPDATE ON THE STATUS OF THE SCIENCE .' xviii
THE RESEARCH STRATEGY USED BY THE WRP ,.. ....xviii
ACKNOWLEDGEMENTS - x'x
CHAPTER 1 • v»" 1
TERMS USED •••• 2
' THE WRP APPROACH AND ITS APPLICATIONS 2
KEY CONCEPTS ...•— • 3
Populations.. 3
Setting —-. 5
Performance Curves • 5
Indicators 7
SUMMARY 8
CHAPTER 2......J 11
MINIMUM INFORMATION NEEDED : 12
FEATURES OF EPA's PERMIT TRACKING SYSTEM... 13
INCORPORATING ADDITIONAL INFORMATION ./. '. .15
• .REPORTING THE INFORMATION 17
SUMMARY •. 19
CHAPTER 3 •— -W ,.-• 23
DECIDING ON A SAMPLING STRATEGY 23
IDENTIFYING PRIORITY AREAS : ; 24
SELECTING SITES 26
Defining the Population of Wetland Projects to Sample 26
An Approach to Improving Decision Making in Wetland Restoration and Creation
ix
-------
CONTENTS
. Defining the Boundaries of a Study Area. • 33
Taking a regional perspective • 33
. Considering ecological setting.... 33
Defining and Sampling the*Population of Natural Wetlands 36
. Finalizing the List of Projects and Natural Wetlands to be Sampled ..36
SUMMARY •—••• 3
CHAPTER 4 .-. •• - • • 43
•. DOCUMENTATION OF AS-BUILT CONDITIONS .'. • 44
Rationale •• • 44
What To Include .v... : 52
ROUTINE ASSESSMENTS - • 56
Rationale :......... : • • 56
. What To Include • 57
COMPREHENSIVE-ASSESSMENTS • 59
• . Rationale •-••.-• 59
What To Include.....;..... 60
ASSESSMENT VARIABLES'.../.... ' 61
' General Information « • 61
Morphometry ....; ••—v -•••• ^
Hydrology • •• 63
Substrate • 63
. Vegetation • •'••••64
: Fauna • • 66
Water Quality ..:... -.:.: i - •— 67
Additional Information...... • 67
DEVELOPING AN EFFICIENT SAMPLING STRATEGY 68
• Data Quality j..j............... 69
Where To Collect Samples - •••• 70
How Many Samples To Collect., :..... • 70
. . When To.Collect Samples.; -71
Controlling Damage To The Site 7^
. SUMMARY .....: •.,.' — 71
i , i
VOLUNTEERS AND NATURAL RESOURCE MONITORING 73
CHAPTERS 87
SUGGESTED WAYS TO REPRESENT THE DATA COLLECTED 87
Performance Curves • 88
-------
CONTENTS
Summary or Descriptive Graphs...... »..:.... 92
Time Series Graphs • —•• 93
Characterization Curves •••• • 93
\ TECHNIQUES FOR DETERMINING DIFFERENCES IN SAMPLES : 96
EVALUATING PROJECTS AND SETTING PERFORMANCE CRITERIA 98
An Extension of the Example ...~.,..105
Example of How to Use Time Series Graphs 105
Example of How to Use Characterization Curves :.1Q7
. SUMMARY , • ;- — • 1°8
CHAPTERS - ' -••• ••••• 111
WETLAND TYPE : • • 111
Determine if the Project is Typical of Wetlands in the Region 112
Influence of Bank Slopes on Wetland Type ....... 112
Relationship Between Bank Slopes and Wetland Area 114
Determine how much land will be required ......... 116
Design when adequate land is available 117
Design when land area available is limited 117
VEGETATION....... • Hg
Example from the Oregon Study - 11-9
Example from the Florida Study • 120
Guidelines for Revegetation of Wetland Projects T 120
, To Plant or Not To Plant? '..'..'..".'...'. •.-• 120
. Generating a Planting List —• 123
What species commonly occur on wetlands in the area?.: :123
Which species are commercially available? 124
Narrow the list of species to generate a planting list ;...124
OTHER IMPORTANT STRUCTURAL CHARACTERISTICS •. 127
Hydrology • 127
Soils/Substrates •••• • - 130
SUMMARY... • • • 131
REFERENCES • 13S
An Approach to Improving Decision Making in Wetland Restoration and Creation
xi
-------
LIST OF TABLES
Table I. Recent publications on wetland restoration and creation xxiii
Table 2-1. Summary of the Section 404 permit databases compiled
by EPA's Wetlands Research Program •• .12
Table 2-2. Minimum categories of data on impacted and
compensatory (created, enhanced, preserved, or restored)
'wetlands recommended for inclusion in a database and data
categories found in EPA's Permit Tracking System (PTS) 14
Table 3-1. Numbers of freshwater mitigation projects in Portland,
Oregon,' by wetland type and size required in Section 404
permits issued by the U.S. Army Corps of Engineers and
.the Oregon Division of State Lands from January 1987
through January .1991 -• 32
Table 4-1. Rationale and uses of variables measured in as-built,
routine, and comprehensive assessments of wetland
projects and natural wetlands..................... —45
Table 4-2. Methods recommended for measuring variables in as-built,
routine, and comprehensive assessments of wetland : .,
•" ' i •••.•••" -A a
projects ^°
Table 6-1. Partial list from which to choose species for planting
on created/restored wetlands in the Willamette Valley,
Oregon.... .......126
Table 6-2. The hydrology planned for created wetlands studied in the
Portland, Oregon metropolitan area in 1987 •••'l28
• :-'•'. " ' . • • ! '{.-.<
Table 6-3. A summary of the findings of recent studies of groups of
wetland projects • .' :"t33'
xn
-------
LIST OF FIGURES
Figure 1 -1 . The steps in the WRP Approach for using quantitative
information to support decision making ........... ..... ..................... 4
Figure 1 -2. Hypothetical performance curve illustrating the
comparison of natural wetlands and projects (in this case
restored wetlands) of the same type and similar size in the
same land use setting relative to a measure of wetland
function... ......................................... • ....... — •• ............................ ^
Figure 2-1 . Examples of the query and results screens from the
Permit Tracking System (PTS) [[[ 1 6
Figure 2-2. Comparison by state of the percent of the Section 404
permits requiring compensatory mitigation that specified
monitoring the project with at least one site visit. ................ .'...18
Figure 2-3. Comparison by state of the net change in area of palustrine
forested wetlands and palustrine emergent wetlands involved
in Section 404 permits requiring compensatory mitigation
over the time period analyzed (see Table 2-1) .... ................... -18
Figure 3-1. -Hypothetical performance"curves.
.25
Figure 3-2. Locations of wetland impacts and creations in Oregon
that occurred between January 1977 and January 1987 27
Fisure3-3. Patterns of Section 404 permitting in California and
• no
Louisiana • • • ..... ^o
Figure 3-4. An example of a form that can be used to compile
information on wetland projects. ......................30
Figure 3-5. Example of a typical mitigation project sampled in the
Oregon Study 32
Figure 3-6. An example of how a U.S. Geological Survey topographic
map can be used to identify subregions 34
-------
LIST OF FIGURES
Figure 3-7. Example of a completed form that can be used during a
field reconnaissance to collect information on potential
study sites ...... . [[[ • ...... •••38
Figure 4-1 . Example of a Field Map to document as-built conditions
of a wetland project ......................... . ....................................... 53
Figure 4-2. Profile of a wetland portraying as-built or current conditions
based on elevational measurements from Basin Morphometry
Transects (BMT1 & BMT2). Transects must match with those
shown on maps with aerial views (See Figure 4-1) ......... . ......... 54
! 'i
Figure 4-3. Map enlarged from U.S. Geological Survey quadrangle
showing drainage area, surrounding land-use, and wetland
location ......... . .............. . ....... .................. • ...................... • ......... 55
Figure 4-4. Example of a Field Map to document conditions found
during a routine assessment as compared to the as-built
condition. Heavy dark line indicates most recent wetland
perimeter and separates areas of dominant vegetation types.
Note change in wetland shape as compared to as-built
conditions shown on Figure 4-1 .................................... . ........ 58
Figure 4-5. Field crew members taking elevation measurements along
a transect ..................................... . .......... • ............................... -62
i i
Figure 4-6. Field crew members using a Munsell color chart to determine
soil hue, value, and chroma [[[ 64
' ."*•.'
Figure 4-7. Botanist reading a vegetation quadrat ..................................... 65
Figure 4-8. Field crew member collecting invertebrates from an
emergence trap ....................................... ... ............................. 67
Figure 5-1 . Performance curve generated using the mean percent organic
matter in the upper five-cm of soil ........................................... 89
'ii,
-------
LIST OF FIGURES
Figure 5-3. Performance curves generated using plant diversity data .;.......9T
Figure 5-4. Hypothetical performance curves illustrating four different
patterns of project development that could be used in making
management decisions ........ ........'........................,.. ................. 93
Figure 5-5. Examples of summary or descriptive graphs ....... ..... ...... ...........94
Figure 5-6. Monthly water levels (cm) for a pair of the created and
' natural wetlands......... .............. .;............... ..... »••••— ................. ?5
Figure 5-7. Example of hypothetical characterization curve.... ......... ..........95
Figure 5-8. Mean percent cover for created and natural wetlands from
the Oregon Study plotted versus project age......... ................... 99
Figure 5-9 Box and whisker plot of cover data for created and natural '
: wetlands....:........:............:;....... ..... . ........ :..;:...«...: ................ 101
Figure 5-1 0. Performance cu'ryfe of plaht;diversity data .....;......... ........... :...1 02
Figure 5-11: Bar graph of the percent of spec'ies overlap between individual
created and natural wetlands ........ .... ........ ............... ..... . ....... ..1 03
Figure 5-1 2. Weighted average scores (Wentworth et al. 1 988) for the
type of vegetation found on individual created (C) and
'"•': natural (N) ' wetlands ..........:..l.'...'v».'...'v ...... .'.'.....yi..V..l ..... . ..... :.'.T04
Figure 5-1 3. Example of an emergent marsh in the Connecticut Study.. ..... 1 1 06
Figure 5-1 4. Example of a pond with a fringe of emergent vegetation
from the Florida Study ...i:..:.. ............ .....;............... ........ .'........106
Figure 5-1 5. Characterization curve of percent organic matter ... ...... .........1 07
Figure 6-1 . Pictures of typical natural (a), and created (b) wetlands in
theOregonl .................. ........................... ............. ...... ..........115
An Approach to Improving Decision Making in Wetland Restoration and Creatit
XV
ion
-------
LIST OF FIGURES
Figure 6-2. Topographical profiles for typical natural (a) and created
(b) wetlands in the Oregon 116
Figure 6-3. Illustration of how to determine the amount of land needed
for creating a wetland given the bank slopes and the depth
from the ground surface to the water table ...118
Figure 6-4. Erosion occurring on steep unvegetated banks at a created
wetland sampled in the Oregon Study— -.119
Figure 6-5. Comparison of the number (a) and the percent cover
(b) of species found on created wetlands in the Florida
Study 121
XVI
-------
BACKGROUND AND ACKNOWLEDGEMENTS
The use of restoration and creation of wetlands to mitigate for permittee
losses and to enhance the wetland resource requires an evaluation of the effi-
cacy of the practice. The key question is: Do created and restored wetlands
develop the same ecological, functions as natural wetlands? In an effort thai
preceded this document and influenced the research that is described in the
following chapters, the Environmental Protection Agency's (EPA) Wetlands Re-
search Program (WRP) assembled a team of experts to compile and documen-
the status of the science on wetland creation and restoration. The resulting
publication, Wetland Creation and Restoration: The Status of the Scienct
(Kusler and Kentula 1990a), took a national view and built on previous wort
in the field/ This book will build on the major findings of that documen
which were: •
(1) Practical experience and available information vary by wetland type
ecological function, and region of the country. The mos
quantitative and best documented information is available fo
Atlantic coastal wetlands. Fewer projects have been implemente<
on the Gulf and Pacific coasts and, correspondingly, there is les:
information. Much less is known about restoring or creating inlanc
wetlands.
(2) Most wetland^restoration and creation projects do not have specifiec
goals, complicating efforts to evaluate "success".' Success is ofter
rated on compliance with permit requirements or establishment o
vegetation. Such measures/however, do not indicate that a projec
is functioning properly or that it will persist over time.
(3) Monitoring of wetland restoration and creation projects has been i
common. Monitoring of sites and comparisons with natural wetland:
over time would provide a variety of information including hov
projects develop over time and howJhey compare with natura
wetlands in the region (Kusler and Kentula 1990b).;
Complementing and including the work presented in Wetland Creatioi
and Restoration, the U.S. Fish and Wildlife Service (FWS) is maintaining th<
Wetland Creation/Restoration database to provide a state-of-the-knowledge re
source based on the published literature. A hard copy of the bibliographi<
material contained in the digital database has also been produced (Schneller
McDonald et al. 1989). "
An Approach toTmprovirig Decision Making in Wetland Restoration and Creation
xvii
-------
BACKGROUND AND ACKNOWLEDGEMENTS
UPDATE ON THE STATUS OF THE SCIENCE
Reporting on wetland creation and restoration has burgeoned in recent
years. Books, manuals, reports, and journal articles have been published on
project design, evaluations of projects, and approaches to monitoring. Table I
highlights recent, publications. j ,
Interest in restoration ecology^ has also flourished in the past five years.
The Society for Ecological Restoration (SER) was established in 1988 and held
its first annual meeting in January 1989 (Hughes and Bonnicksen 1990). As of
April 1992, membership in the Society had grown to over 1700. Indicative of
the demand for'information on Ecological restoration, SER is now in the
process of establishing a journal/Restoration Ecology, to accompany its
newsletter and its periodical, Restoration and Management Notes.
Three recent compendiums on;environmental restoration deserve mention.
In Restoration Ecology: A Synthetic Approach to Ecological Research (Jordan
et al. 1987) some two dozen ecologists discuss the heuristic or intellectual
value of ecological restoration. Specifically, they address ecological restora-
tion as a way of raising basic questions and testing fundamental hypotheses
about the communities and ecosystems being restored, i.e., as a technique for
basic research. Rehabilitating Damaged Ecosystems (Cairns 1988) takes a
broad, eclectic view. Authors representing diverse fields present case histories
from a variety of systems, in addition to discussions of planning procedures
and approaches to management. Finally, Environmental Restoration: Science
and Strategies for Restoring the Earth (Berger 1989) reports the results of the
1988 Restoring the Earth Conference, presenting an overview of the most cur-
rent techniques and processes for restoration, discussions of current issues, and
descriptions of restorations of assorted systems.
THE RESEARCH STRATEGY USED BY THE WRP
This document synthesizes the results of over five years of research by the
WRP to illustrate and support an ^approach to evaluating wetland restoration
and creation. The strategic plan for the research (Zedler and Kentula 1986)
recommended that existing mitigation projects be treated as "experiments in
progress". Implementation of the strategy led to the theme of this document—
we must learn from what we have done and use the information to improve fu-
ture decisions. ;
Several studies produced the information on which this document is
based.' These studies are grouped under the two lines of research implement-
ed: examination of the patterns and trends in permitting under Section 404 of
XVIII
-------
BACKGROUND AND ACKNOWLEDGEMENTS
the Clean Water Act, and evaluation of freshwater mitigation projects. The re-
ports and papers recounting the results are cited throughout this document
We analyzed portions of the Section 404 permit records from different re-
gions of the country to determine patterns and trends in permitting activity aric
to document the cumulative effects of the associated management decisions
on the resource. Results from these and similar studies can be used to evalu-
ate wetland management practices, especially the use of compensatory mitiga-
tion. We also conducted a number of pilot studies to compare characteristic:
of mitigation projects and natural wetlands, to test approaches and method:
for collecting data, and to evaluate project design and compliance with perrrir
conditions. In particular, data from the pilot studies of freshwater mitigatior
projects in Portland, Oregon; Tampa, Florida; and Connecticut were used
The experiences resulting from the study of a created wetland in Seaside, Ore
gon, are reported in the special insert, Volunteers and Natural Resource Moni
toring. The projects examined in all studies were typically less than or equa
to five years old, and the majority were what is probably the most comrhor
freshwater mitigation project nationally—a pond with a fringe of emergen
marsh, in the Oregon Study a group of 11 created wetlands were comparec
with a group of 12 natural wetlands. In this case the entire population of miti
gation projects existing at the time was sampled. In the Florida Study a grou[:
of nine created wetlands were compared to a group of nine natural wetlands
In the Connecticut Study, five Connecticut:pepartrrient of Transportation miti
gation projects were paired with five natural wetlands. In both the Florida'anc
Connecticut studies only a subset of the projects in the area were sampled. •
The general framework for each study was similar and was provided b\
the WRP scientists. It included sampling design, methods, quality assurance
procedures, and guidelines for data analysis. This was done so that the frame
work could be evaluated and the findings from all four studies could be com
pared, fn addition, the principal investigators of each study provided a cri
tique of the framework and introduced new components into their respective
studies.-,- '" "" "" " """..-::„"'.:.:_.: ~:.. " :_. ". : •'" -
ACKNOWLEDGEMENTS i
We appreciate the contributions made by many individuals during th<
prepiration of this do^cument. Personnel from the Wetlands Division of EPA':
Office of Wetlands, Oceans, and Watersheds have provided constructive, de
tailed review comments to the authors and have been extremely supportive o
the research necessary to, bring this book;to completion. We acknowledge
their responsiveness and realize that their efforts guarantee the WRP is focus
An Approach to Improving Decision Making in Wetland Restoration and Creation
XIX
-------
BACKGROUND AND ACKNOWLEDGEMENTS
ng on issues important to the Agency. We continue to appreciate the person-
il support of John Meagher, Director, Wetlands Division. In particular, we
/yant to thank Doreen Robb, Wetlands Divisions' liaison to the research pro-
gram, for her efforts to-help us communicate effectively and accurately.
Scientific Contributions to This Document
The research that was synthesized in this document was conducted with
:he help of many people. In the case of the permit studies, Tina Rohm (former-
y with Northrop Services, Inc.) developed a data management system to com-
Dile the data. Her system is the model for the Permit Tracking System, which
,vas designed and programmed by Robert Gibson (ManTech Environmental
Technology, Inc.) of the WRP. Robert used his creativity in computer pro-
gramming to simplify and expedite data entry and analysis of the permit infor-
Tiation.
Jim Good (Oregon State University), Kathy Kunz (formerly with EPA-Re-
gion 10), Michael Rylko (EPA-Region 10), Jane Griffith and Sharon Lockhart
formerly with FWS, Laguna Miguel Field Office), Paul Price (Paul Price Associ-
ates, Inc.), and Edwin W. Cake (Gulf Environmental Associates) collected and
sntered the data from Section 404 permit files. Millicent Quammen (FWS, Na-
:ional Wetlands Research Center, Corpus Christ! Field Station) served as the
Droject officer for the studies of permitting in Texas, Arkansas, Louisiana, Mis-
sissippi, and Alabama. The permit studies are discussed in Chapter 2.
The Oregon Study was led and conducted by the authors, who are mem-
oers of the WRP. However, it is important to recognize the contributions of
i/VRP's senior geographer, Brooke Abbruzzese (ManTech Environmental Tech-
nology, Inc.). Ms. Abbruzzese played a major role in developing the funda-
mental approach to site selection. The methods she developed for the Oregon
5tudy (Abbruzzese et al. 1988) became the template for subsequent studies.
Moreover, she has continued to contribute by providing advice as we tested
and refined the methodology. Site selection is discussed in Chapter 3.
In addition we want to acknowledge the work of several interns from the
Geography Department at Oregon State University, Corvallis, Oregon. Jack
Davis and Eric Hughes performed the spatial analysis and prepared the associ-
ated draft maps that resulted in the Section 404 permit maps of Oregon and
Louisiana used in Chapter 3. Tracy Smith developed and prepared the final
/ersions of the Field Maps used in Chapter 4.
Dr. Mark T. Brown (Center for Wetlands, University of Florida,
Gainesville) led the Florida Study. He and his staff developed a system for
sampling a wetland that minimizes the number of times the site is traversed
xx
-------
BACKGROUND AMP ACKNOWLEDGEMENTS
and, therefore, minimizes the damage to the site (Brown 1991). Dr. Browr
also assisted in refining our ideas on site selection. In particular, he developec
the Landscape Development intensity (LDI) index as a way to quantify the
landscape setting of a wetland as defined by the land uses in: the vicinity of the
site (Brown 1991). Use of the LDI is discussed in Chapter 3.
Dr. William A. Niering (Connecticut College, New London) led the Con
necticut Study. He and graduate student Sheri R. Confer contributed anothe
dimension to our studies by sampling over time. The Oregon, Florida, anc
Connecticut Studies sampled the wetlands once during the growing season. -Ir
addition, Confer and Niering measured water levels and observed animal use
monthly for over a year and characterized plant community composition dur
ing two consecutive growing seasons (Confer 1990, Confer and Niering Ir
press). The use of such time series data is discussed in Chapter 5.
Finally, Dr. Milton Weller (Texas A&M University, College Station), ir
conjunction with Dr. Gerald W. Kaufmann (Loras College, Dubuque, Iowa
and Dr. Paul A. Vohs, Jr. (FWS, Iowa Cooperative Fish & Wildlife Researd
Unit/Ames, Iowa) expanded our ideas by providing significant information w<
were not able to collect in .the otheLStudies. They repeated their pre-im
poundment and early post-impoundment studies done almost 30 years earlie
when Drs. Kaufmann and Vohs were graduate students studying with Dr
Weller. These studies documented the changes in the wetlands and the asso
elated waterfowl that resulted:from the4mpoundrnent:in 1961'of Elk Creek, ;
small creek in Worth County, Iowa '(Weller et al. 1991). Information on thi
development of projects over such a long period of time is extremely ran
(Kusler and Kentula 1990b), and, therefore, very valuable. .
Technical Contributions to This Document
"' '•''• The authors have benefited greatly from the suggestions of those who re
viewed this document. The Wetlands Division, EPA Regions 3, 9, and 1C
other agencies, and members of the academic community gave generously c
their time to provide comments. Richard Coleman coordinated a review b
the U.S. Army Corps of Engineers' Wetlands Research Program at the Water
ways'Experiment Station; Susan Haseltine, by the FWS Northern Prairi
Wildlife Research Station; Lee Ischinger, by the FWS National Ecology Re
search Center; and Virginia Van SIckfe-Burkett,.by the FWS National Wetlam
Research Center. Specifically, we thank the following individuals'for the!
thoughtful reviews: Barbara Bedford (Cornell University), Marcia Bollma
(Mantech Environmental Technology, Inc.), Mary M. Davis (Army Corps c
Engineers), Kate Dwire (ManTech Environmental Technology, Inc.), Pat
An Approach to Improving Decision Making in Wetland Restoration and Creation
xxi
-------
BACKGROUND AND ACKNOWLEDGEMENTS
3arrett (Federal Highway Administration), Jerry Grau (Fish and Wildlife Ser-
vice), Randy Gray (Soil Conservation Service), Susan Haseltine (Fish and
Wildlife Service), Hal Kantrud (Fish and Wildlife Service), Tom Kelsch (EPA
Wetlands Division), Dennis King (Maryland Institute for Ecological Econom-
cs)r Russ Lea (North Carolina State University), Anne Marble (A.D. Marble &
Company), Geoffery Matthews (National Oceanic and Atmospheric Adminis-
ration), Thomas Minello (National Oceanic and Atmospheric Administration^
/Villiam Niering (Connecticut College), Philip North (EPA Region W, Alaska
Operations), Philip Oshida (EPA Region 9), Barry Payne (Army Corps of Engi-
neers), Bruce Pugesek (Fish and Wildlife Service), Doreen Robb (EPA Wetlands
Division), Charles Segelquist (Fish and Wildlife Service), Paul Shaffer (Man-
Tech Environmental Technology, Inc.), Bill Sipple"(EPA Wetlands Division), Art
>pingarn (EPA Region 3), Michelle Stevens (Washington Department of Ecolo-
jy), Rich Sumner (EPA Regional Liaison), Curtis Tanner (EPA Region 10), Ron
Futtle (Soil Conservation Service), Virginia Van Sickle-Burkett (Fish and
Wildlife Service), Fred Weinmanri (EPA Region 10), Milton Weller (Texas A&M
University), Joy Zedler (Pacific Estuarine Research Laboratory, San Diego State
University).
Finally, we recognize those who turned our writings into a finished .docu-
ment. This manuscript could nofexist without Kristina Miller's exceptional
ikills in document and graphics production. We especially appreciate her
:heerfu! tolerance of our times of indecision and needs for.immediate
zhanges. Linda Chesnut-Korwin's creativity improved the format of the Execu-
ive Summary, special insert, Volunteers and Natural Resource Monitoring,
ind Tables I, 4-T and 4-2. Graphic'artist, Linda Haygarth produced the figures
hat needed a "human touch", i.e., could not be made on the computer.
XXII
-------
Table I. Recent publications on wetlands. Books, proceedings, and reports pub-
lished since the preparation of Wetland Creation and Restoration: Th«
Status of the Science (Kusler and Kentula 1990a) are listed and describee
below. This list is presented as a guide to the recentliterature'and is no
intended to be comprehensive. Numerous other references are discussec
throughout this document and are listed in the References.
Coastal Marshes: Ecology and Wildlife Management
Chabreck(1988) i
This book describes the coastal marshes of the United States, their form, functions, ecological
relationships, wildlife value, and their management for wildlife. The marshes of the northern
coast of the Gulf of Mexico are emphasized.
The Ecology and Management of Wetlands, Volumes 1 & 2
Hook etal. (1988) . . ...
This book contains the proceedings of a symposium held at the College of Charleston in
1986. The contributions have been organized to focus on (1) the resource and the basic biol-
ogy and ecology of wetland plants, animals, soils, hydrology and their values and interac-
tions, and (2) the practicality of applying such information to protect and manage the wetland
resource.
Wetland Modelling
Mitsch et al. (1988)
This volume is a statement of the state of the art of modelling approaches for the quantitative
study of wetlands. Chapters present different aspects of wetland modelling or a case study
characteristic of wetland modelling. Modelling approaches for a wide variety of wetland
types are included, as well as models with an emphasis on wetland hydrology, biological
productivity and processes and wetland management, and for designing and summarizing
large scale research projects.
Proceedings of In Inferriatiohal Symposium: Wetlands and River Corridor
Management
Kusler and Daly (1989)
The papers in this volume address river and stream corridor management, including the adja-
cent riverine and estuarine wetlands, from a natural systems protection and restoration per-.
spective. Most of the papers were presented at the international Symposium: Wetlands and
River Corridor Management which was held in Charleston, South Carolina, in July 1989.
Wetlands Ecology and Conservation: Emphasis in Pennsylvania
Majumdar et al. (1989)
Wetland experts in the field address a variety of topics on geologic, chemical and biological
aspects of wetland ecology. Several chapters are devoted to wetland preservation, and also
to increasing our wetland resources through enhancement and mitigation. Important wetland
issues such as endangered species/mitigation, and pollution abatement are discussed in de-
tail. The book explains the complexities of protecting wetlands, from delineating boundaries
to applying for a permit, to restoring a degraded wetland.
Freshwater Wetlands and Wildlife
:':_..,. .- - : SharitzandGibbons(1989)
This volume is a product of the Freshwater Wetlands and Wildlfe Symposium held in
Charleston, South Carolina, in March' 1986. It addresses issues related to natural, man-man-
aged, and degraded ecosystems. The first section deals with the functions and values of wet-
lands, including their use as habitat, role in trophic dynamics, and basic processes. The sec-
ond section discusses their status and management,.including techniques for assessing value,
laws for protection, and plans for management.
"Ah Approach to Improving Decision Making in Wetland Restoration and Creation
xxiii
-------
Table I. (continued)
Northern Prairie Wetlands
van derValk (1989)
This book is primarily a review of the ecology of palustrine and lacustrine wetlands in the
northern prairie region, i.e, the prairie pothole region of the United States and Canada plus the
». i i._r_-i_jLMi_ 1^ j~.,~i«.~~i ™.t ~t = cwmr><->cinm hplH in 1985 at the Northern Prairie
Nebraska sandhills. It developed out of
Wildlife Research Center of the U.S. Fish
a symposium held in 1985 at the Northern Prairie
ind Wildlife Service in Jamestown, North Dakota.
Buffer Zones for Water, Well Jmds, and Wi Idlfe in East Central Florida
Brojvnetal. (1990)
This report presents a method for estimating buffer sizes necessary in counties in east central
Florida to achieve wetland protection through minimization of groundwater drawdown in wet-
lands, minimization of sediment transport into wetlands, and protection of wildlife habitat.
Standards and criteria, minimum buffer requirements, and site-specific measurements that
could be used to determine buffers on a s te-by-site basis are proposed.
I
Ecological Processes and Cumu ative Impacts: Illustrated by Bottomland
Hardwoodj Wetland Ecosystems
Gossfel ink etal. (1990)
This book presents the results of three workshops convened by the EPA and facilitated by a
~ -•-• •- • " ' Center of the U.S. Fish and Wildlife Service to solicit
team from the National Ecology Research
expert advice on bottomland hardwood
are summarized in chapters on hydrolo
..
forest ecosystems. The reports from the workshops
soils, water quality, vegetation, fisheries, wildlife,
are summan^cu in *-iidpi.ci:» un IIJTUIWH-/HJT, awfi.*, »»«!.*-• ^MU...,, --0 / ----
ecosystems processes and cumulative impacts, and culture/recreation/economics.
Forested Wetlands
Lu4o etal. (1990)
This volume of Elsevier's series, Ecosysterhs of the World, is intended as an introduction to the
subject of forested wetlands. The first part reviews available information on the structure and
function of forested wetlands and is strongly biased toward forested wetlands in the Caribbean
and the United States. The second part p resents case studies and descriptions of forested wet-
lands from other parts of the world.
A Guide to Wetland Functional Design
Marble (1990)
design guidance that was developed by working each of the function
Technique (WET) backwards to identify the predictors which
discusses conceptual design, site selection, and site
This guidebook presents
keys in the Wetland Evaluation
generate a "high" rating. The guidebook
design.
and
A Manual for Assessing Restored
from
Pacific Estuarine
The manual summarizes reference data
case study from San Diego Bay, and promotes
restored and natural wetlands by
niques.
Synthesis of Soil-Plant Correspond
Throughout
This report synthesizes the information co
ries of 12 studies designed to describe J
located in 11 states throughout the U.S.
Natural Coastal Wetlands, With Examples
Southern California
Research Laboratory (1990)
from several Southern California wetlands, presents a
the standardization of assessment methods for
recorrjmending specific sampling and measurement tech-
lence Data From Twelve Wetland Studies
the United States
Segelbuistetal. (1990)
illected for the U.S. Fish and Wildlife Service in a se-
relation between soils and vegetation in wetlands
the
XXIV
-------
Table I. (continued)
Mitigation Site Type Classification System (MIST)
White etal. (19i90)
A system for evaluating sites to be restored as forested wetlands based upon the Condition of
' . - IK f __._•*«_. n_——™ *.( frvmnrf-j-irinrr resnt lirorl 1C \{&\ff*n to trIP <5ltP_
key site factors controlling productivity. Degree of
classification.
Wetlands of North
monitoring required Is keyed to the site
America
)
Niering(199', . .
This beautifully illustrated book provides an introduction to wetlands. It is divided into four
chapters representing each of the major wetland typ<|s—freshwater marshes, coastal wetlands,
swamps and riparian wetlands, and bogs and peatlan is.
Estuarine Habitat Assessment Protocol
Simenstad etal. (
The Protocol was developed in response to the neec
the function of estuarine wetlands and associated ne
based on the use of systematic, on-site measurement^
utes of the habitats identified as being functionally irr
for procedures that quantitatively assess
rshore habitats for fish and vfildlife. It is
af habitat function by assessing the attrib-
portant to fish and wijdlife.
neirshore
Freshwater Wetlands
Creating
Hammer (19S
This book is an attempt to organize and present ii
freshwater wetlands accumulated by wetland scientis
2)
information on methods to create or restore,
ts and managers during the last 50 years.
Restoration of Aquatic Ecosystems: Scienc
National Research Cot
This volume examines the prospects for repairing th<
aquatic resources: lakes, rivers and streams, and w<
aquatic restoration, with practical recommendatib is
scale of projects and needed governmental action." -
ties throughout the country are presented.
;, Technology, and Public ^Policy
Council (1992) -. ,
damage society has done toithe nation's
we|tlands. It outlines a national; strategy for
._ covering both the desired scope and
Ease studies of aquatic restoration activir
Chapter 13: Wetland Restoration,
U.S.DA. Soil Conservatior
Wetland restoration, enhancement, and creation
Conservation Service's Field Handbook. This chapte
that formerly described the construction of dikes and
Enhancement, and Creation
Service (1992)
is now the focus of Chapter 13 of the Soil
replaces and expands upon jthe material
levees.
An Approach to Improving Decision Making in Wetland Restoration and Creation
XXV
-------
-------
CHAPTER 1
Introduction
The management of the Nation's wetland resource is characterized by
controversy. There is agreement that wetlands are an important component of
the landscape, providing a variety of ecological, social, and aesthetic benefits.
There is also agreement that over half of the resource has been lost due to con-
version to other uses (Tiner 1984), and that the resource, needs protection.
There is, however, disagreement as to how wetland protection Is to be accom-
plished.
Efforts to protect wetlands are increasing as are the economic pressures to
convert them. Government agencies attempt to balance the needs for protec-
tion and development through their management decisions as to when and
where wetland alteration will be permitted and compensatory mitigation will
be required. At the same time, parties immersed in wetlands issues expect
some level of predictability in wetland management and regulatory decisions.
To establish consistency in the management process, we need a coherent
framework for linking current wetland management and regulation. Creative
new initiatives are also necessary to address significant voids in protection, in-
efficiencies in existing programs, and counterproductive public actions and in-
centives (The Conservation Foundation 1988).
A comprehensive program of wetland management and^regulation re-
quires information on the ecological functions of wetlands individually and in
the local landscape, and on our ability to create and restore wetlands. Re-
search projects implemented by the Environmental Protection Agency's (EPA)
Wetlands Research Program (WRP) are designed to supply this information
Chapter 1: Introduction
-------
Zedler and Kentula 1986, Leibowitz et al. 1992). Specifically, Agency per-
onnel surveyed in the planning process agreed that the key question.is: Do
estored and created wetlands perform the same ecological functions as natur-
il wetlands? Moreover, efforts to evaluate the success of wetland restoration
md creation projects have been complicated by a lack of stated project goals
ind by a lack of agreement on what constitutes success. To begin the process
rf defining success in an ecologically meaningful way, we have developed an
ipproach to establishing ecological criteria for wetland restoration and cre-
ition based on results of our studies and related current research. The WRP
\pproach to evaluating wetlands and wetland projects can be used to tailor re-
source management to meet specific local and regional needs, , ,
TERMS USED ' ; ' ' r,
The WRP Approach can be used to guide the evaluation of created, re-
stored, enhanced, rehabilitated, constructed, and, fundamentally, any human-
•nanipulated wetlands, as well as natural wetlands. To avoid listing all possi-
ole situations that might apply and to avoid the current problems of multiple
meanings for some terms, we use projects to refer to all wetlands that were
created or were "improved" by human activities for a specific purpose (e.g.,
mitigation). We use natural to refer to wetlands that occur naturally in the
landscape. However, there are times when we need to be more specific. In
those cases we use restored to refer to any manipulation of a site that contains
or has contained a wetland and created to refer to attempts to construct a wet-
land in an area that never has contained a wetland. Population, statistically
speaking, refers to those wetlands of a similar type and size, either natural or
projects, occurring within a geographically defined area. : -
As mentioned above, the word success has a number of meanings as it re-
lates to wetland projects. Success of projects is often rated either on the basis
of compliance with permit requirements, or on the basis of whether or not the
projects were implemented (Quammen 1986). We believe that success must
be defined in terms of the project objectives, i.e., what is acceptable for a par-
ticular project in a, specific locale. For some, success is meeting the terms of
the contract; for others, replacement of all aspects of a natural system; for oth-
ers, replacement of some functions to some level. We leave the definition of
project objectives, and the associated success, to those planning or regulating
the project. Instead, we offer a way to quantify ecological performance and,
ultimately, verify that project objectives have been met, however they are de-
fined. . ,
THE WRP APPROACH AND ITS APPLICATIONS
The WRP Approach demonstrates how information from a monitoring pro-
gram that includes both natural wetlands and those restored and created can
An Approach to Improving Decision Making in Wetland Restoration and Creation
2 , ,
-------
be used to develop performance criteria and suggest Improvements to the de-
sign of future wetland projects. The same information can be used to evaluate
the effectiveness of the management strategy being used. Ultimately, informa-
tion collected over time can be used to evaluate and refine performance crite-
ria and design guidelines, and to ascertain when a project is developing as ex-
pected and when corrections are necessary.
Although wetland restoration and creation are central to this document, it
is not a step-by-step approach to building wetlands. Instead, our-Approach is
a framework for the development of ecologically defensible management
strategies for restoration and creation that are tailored to local and regional
needs. Such a framework would assist managers in identifying which projects
proposals and permit applications should 1) be most critically reviewed be-
cause of low probability of success; 2) require the most comprehensive checks
on project design and implementation because of uncertain probability of suc-
cess; and 3) require minimal checks on design and implementation of the miti-
gation project because of high probability of success. The overall framework
is flexible enough to be applied in any area and to any wetland type. Howev-
er, the monitoring techniques and examples presented will transfer.most readi-
ly to freshwater nontidal wetlands because they have been the WRP's focus.
KEY CONCEPTS
The steps in the WRP Approach are diagrammed in Figure 1-1. Each of
the following chapters discusses one or more of the steps and illustrates the
concept using data from the WRP's field studies of natural wetlands and miti-
gatioWprojects."6rfe'fIy,,the Approach prescribes compiIi ng and anal yzi ng i n-
formation from the project files. The results of the analysis are then used to se-
lect sites for inclusion in a monitoring program. Data collected are analyzed
to determine the performance of the projects, and generate performance crite-
ria and design guidelines for future projects.
Populations
The overall strategy centers around comparisons of samples of populations
of natural wetlands and projects. We feel that it, is_,irnportantjtq_consider the
variability of ecosystems when making ^management decisions,, especially
when setting criteria for performance. Case studies of single sites, and com-
parisons of pairs of sites do not provide information that can be extrapolated
with known certainty to the population as a whole. Therefore, we use the
term population in a statistical sense.
The samples of both populations being compared are used as reference
sites, i.e., the natural wetlands and wetland projects being sampled are each a
group of reference sites. The natural wetlands_are_a reference against which
the development of the projects is judged." The" older projects are a reference
,.„-.. :.-•--.-— - - Chapter 1: Introductior
-------
(5)
(4)
(5)
USING EXISTING INFORMATION (2)
SETTING PRIORITIES/
SELECTING SITES (3)
MONITORING (4)
EVALUATING THE DATA (5)
DEVELOPING PERFORMANCE
CRITERIA (5)
(6)
(6)
IMPROVING DESIGN
GUIDELINES (6)
INFLUENCING FUTURE
DECISION - MAKING
Figure 1 -1. The steps in the WRP Approach for using quantitative information to support .
decision making; in particular, for developing performance criteria and evaluating
design guidelines for restored and created wetlands. The numbers in parentheses '
indicate the chapter in which the concept diagrammed is discussed.
An Approach to Improving Decision Making in Wetland Restoration and Creation
4
-------
against which the development of similar, newer projects is judged. Both are
important. The natural wetlands are used to establish how well goals are
being met; the older projects are used to verify, that other projects are develop-
ing as expected or to detect the results of changes in design.. , .
Setting
The ecological setting of the wetlands is considered in defining the popu-
lations and stratifying the samples (see Chapter 3 for details). Brooks and
Hughes (1 988) suggested Omernik's (1 987) ecoregions as a framework for the
selection of reference wetlands because they reflect regional patterns of land
use, land surface form, potential natural vegetation, and soils. Moreover, they
were shown to be an appropriate framework for the selection of reference sites
in studies of streams in Arkansas (Rohrn et al. 1 987), Ohio (Larsen et al. 1 986,
Whittier et al. 1987), Oregon (Hughes et al. 1987, Whittier et al. 1988), Col-
orado (Gallant et al. 1989) and Wisconsin (Lyons 1989). Because wetlands
and their functions are affected by many of the same factors important to the
function and quality of streams, we adopted Omernik's (1987) ecoregions as
the regional framework for the selection of sites. In addition, we account for
potential effects of land use and position in a watershed in site selection by
grouping wetlands in similar land use settings and watershed positions. Com-
parison of projects with natural wetlands occupying similar landscapes and,
therefore, having potentially similar ecological conditions, ensures that what is
expected of a wetland project is within the bounds of possible performance
given the setting. This framework follows a rationale previously outlined by
Performance Curves :
A key analytical tool of the Approach is the performance curve. The per-
formance curve documents the development of the ecological function of pro-
jects over time relative to levels of function of similar natural wetlands. Figure
1-2 illustrates one form that a curve could take in an idealized, hypothetical
example. Fundamentally, the ability to replace wetland function and the way
in which that replacement occurs depend on the type of wetland, the function
to be restored or created, and, in the case of restoration, the type of impact
that altered the original wetland (Kusler and Kentula 1 990b).
Key aspects of the performance curve are labelled on Figure 1-2 which il-
lustrates one possible scenario, in this ca^e 'a restoration. A is the mean level
of wetland function of the restored sites prior to the implementation of the pro-
ject. A>0 in the case of restoration where there is some level of wetland func-
tion prior to the project and A=0 in the case of a creation where there is no
level of wetland function prior to the project. B is the mean level of function
after the restored wetlands have fully, matured. The difference between A and
Chapter 1: Introduction
-------
i -
1 •
LU
1
O
1
LU >
O
z
«=r i
FICAL PERFORIV
' T T T .
LU '
O
iT* 1
> 1
I
1 • 1
1 • 1
1 .
1 •
1
1
I1 •
1 •
H I-
• 1
• 1
1
1
*— 1
-. — 1
- 1
. 1
1
••
, i
1 i
V,
Y
\
uojjounj j.o
• .. ' , .. • = • , . ' •
Legend
• = mean for natural and restored'
. wetlands
£ = ± 1 standard error • •
r = Natural Wetlands
— = Restored Wetlands
,„ .
O
X
Sv
::. '•'••• Ss^V'
I i .1 . 1 T n
9jnseai/\j
i i i
iis case restored wetlands) of the
A is the mean level of function
:eded for the projects to mature, i.e
the time monitored.
i3 «= •*= •"
o -3. v ^
'§•-0 •£ J2
Q. C en flj
_- JS "™" >
tJ T^J f ) p-
— TO > "fd
"i-*5 1 ^
CS) | 1 e •£
— r™ QJ QJ **—
1 1 1 1 1
Year of Monitor!
5 the comparison of natura
d use setting relative to a rr
function of the projects wl
ie mean level of function <
•£ g o z
1 || 5
= gj-^ «5
3 c £!
0) 4> •£ O
S '* " =
i ^ ". I
•S -i § §
Q.-Q P3 >~
'•§ a. £! 5
•£. •S" S oi
irll..°
cji
•• t—
-------
£, therefore, represents the amount of functijon gained due to the construction
Of the projects. C is the time needed for projects to mature and reach a stable
level of function. D is the mean level of function of the natural wetlands over
the time monitored. D minus B is the difference in the mean level of function
of the natural wetlands and that of mature r. rejects. The values of A, B, C, D
and D minus B, as well as the shape of the curve, provide information relevant
to wetland management. The shape of the curve can be used to decide when
to monitor. For example, inflections in the curve would indicate a; change in
the rate of development, so monitoring immediately before and after you ex-
pect the change could be used to see if the change occurred as expected. The
curve can also be used to decide when a project has met or should meet its
goals. For example, this could be when the project is C years old, OR has a
level of function equal to B plus or minus one standard error of the mean, OR
passes a critical stage in development, i.e.,
experience has demonstrated that
projects reaching a certain point on the curvi; mature as expected.
Using a similar approach, research carried out by EPA's Office of Policy,
Planning, and Evaluation in cooperation with the University of Maryland's
Center for Environmental and Estuarine Studies focused on an Integrated
framework for evaluating the cost and performance of wetland creation and
restoration projects (King 1991 a and King 1991 b). King (1991 a) demonstrates
how the shape of the performance curve for
a given site can be affected by the
characteristics of the creation or restoration project which are often deter-
mined by the amount of resources commi
.ted to the project. In follow-up
work, King (1.991 b) argues that financial incentives in wetland mitigation mar-
kets reward low cost, not high quality wetland restoration, and account for the
relatively poor performance of many restorat
Indicators
Indicators, generally speaking, are vari
ticular wetland functions that their presence
istence or level of function. Generation
having reliable indicators of wetland function
used witlrany regularity, they need to be
tional performance in a reasonable amount
dollars and in damage to the wetland.
morphology or species present, are readily c
requirements of expediency and economy
Therefore, measures of structure are frequen
function. Some of the typical measures of
measured over time, measures of function.
When measured once, diversity describes
terest at one point in time and is a measure
oftre
Measures
than
the
on projects.
ables so closely associated with par-
or value is symptomatic of the ex-
performance curves depends on
If the indicators are going to be
sensitive enough to determine func-
qf time at a reasonable cost both in
of wetland structure, e.g., site
vailable and more often meet the
do direct measures of function.
itly used as indicators of wetland
wetland structure become, when
An example of'this is diversity.
community of organisms of in-
structure. When measured over
of:
Chapter 1: Introduction
-------
ime, it documents the system's ability to maintain a level of diversity, which is
i function. ..
Indicator development in wetland science has focused primarily on van-
ibles that signify a wetland is present, not specific wetland functions. Ai-
:hough verifying that wetland function exists is the ultimate goal, it is impor-
:antto establish that a project is, indeed, a wetland and that it maintains the
iaracteristics of a wetland over time. Therefore, we recommend that at least
Dne variable measuring each of the three parameters (wetland hydrology, hy-
drophytes, and hydric soils) that indicate the presence of a wetland be includ-
=d in any monitoring program. At the minimum, you will be establishing that
the wetland functions associated with a particular type of wetland may exist at
some level, since the characteristics of that wetland are present.
We envision that a set of performance curves will be produced over time
for each indicator or function measured. What is measured is determined by
the goals of the resource management program arid the specific projects
(Chapter 4) The curves are then examined to identify patterns that can be
used as performance criteria, to track a.project's development (Chapter 5), and
to improve project design (Chapter 6).
SUMMARY , . . r
The WRP Approach is a framework for collecting and using information
on populations of restored and created wetlands in a given locale. By building
a quantitative database of information on what works, what constitutes a suc-
cessful wetland project, what does not work, what causes damage or loss, the
Approach provides the scientific information necessary to help resource man-
agers make decisions that will work and that are defensible.
An Approach to Improving Decision Making in Wetland Restoration and Creation
8
-------
(5)
(4)
(5)
's> >
SETTING PRIORITIES/
SELECTING SITES (3)
MONITORING (4)
EVALUATING THE DATA (5)
DEVELOPING PERFORMANCE
CRITERIA (5)
(6)
(6)
IMPROVING DESIGN
GUIDELINES (6)
INFLUENCING FUTURE
DECISION - MAKING
-------
-------
CHAPTER 2
Using Existing Information
A deluge of wetland project files exists in many federal, state, and private
agencies involved with wetland regulation and management. The details of
the final project agreements, however, are seldom documented or accessible.
Decisions typically are made on a case-by-case basis without benefit of quan-
titative information on how previously granted projects relate to the current
proposal or how they affect the status of wetlands in the region (Kentula et al.
1992, Holland and Kentula In press, Sifneos et al. In press(a), Sifneos et al. In
press(b)). Data in the project files, therefore, must be updated, compiled, ana-
lyzed, and reported if the information is to be reflected in management deci-
sions.
The information in the project files, once in an accessible format, can be
used to determine wetland types and locales at risk, to evaluate wetland man-
agement practices, and ultimately to influence policy. For example, analysis
of previous trends in permitting can reveal locales, wetland types, and func-
tions subject to the most intense permitting activity. With knowledge of such
trends, permitting agencies can take actions to avoid additional losses in wet-
land numbers, types, functions, and area.
Although wetlands are constantly being lost to natural forces and human
activities, such as erosion, drainage, and land-clearing, regulation through per-
mitting is one mechanism by which agencies can influence the wetland inven-
tory. Periodic assessments of the cumulative impacts of various permitting sys-
tems, (e.g., Clean W_aterAct_S.ect]ons 404 and 401, Rivers ;and Harbors Act
Section 10, and state regulations) on wetlands are essential for determining the
Chapter 2; Using Existing Information
11
-------
overall effects of permitting on the wetland resource. Unfortunately, the quali-
ty of the documentation of management decisions has been inadequate for re-
liable descriptions of trends in the status of the resource or for evaluation of
management strategies. For example, we analyzed databases containing infor-
mation from portions of the Section 404 permit record from the 1970s and
1980s for eight states (Oregon, Washington, Louisiana, Mississippi, Alabama,
Texas, Arkansas, and California) (Table 2-1). In all eight states, information on
the impacted wetlands and mitigation projects was either lacking or of poor
quality. Approximately 40% of the impacted wetlands and mitigation projects
i,n California lacked acreage data; therefore, area trends reported for the state
might be misleading (Holland and Kentula In press). Furthermore, information
on project completion dates was inadequate for all eight states. In Louisiana,
only 3% of the mitigation projects had completion dates listed in the permit
records (Sifneos et al. In press(a)). A large percentage of permits issued in sev-
eral states lacked specific locations for the wetlands. A better assessment of
the effects of permitting on wetlands would be possible if record keeping were
improved and standardized. In particular, this would allow consideration of
the cumulative effects of individual permit decisions on the wetland resource.
MINIMUM INFORMATION NEEDED
Although detailed information on all permits and projects should be kept
in the files, it is imperative that a subset of this information be compiled, en-
„ ' , '• ' I 'i
Table 2-1. Summary of the Section 404 permit databases compiled by EPA's Wetlands Research
Program. IMP=number of wetlands impacted; COMP=number of compensatory
wetlands. .
State Information Compiled # Permits # Wetlands
IMP CQMP
OR All permits requiring mitigation 58 82 80
1977-January 1987
WA All permits requiring mitigation, 1980-1986 35 72 52
" TX All permits involving freshwater wetlands and 46 71 72
requiring mitigation, 1982-1986
AR All permits involving freshwater wetlands and 7 8 9
requiring mitigation, 1982-1986
AL All permits involving freshwater wetlands, 18 28 23
1982-August1987
MS All permits involving freshwater wetlands, 10 11 6
1982-August1987
LA All permits involving freshwater wetlands, 226 258 116
1982-August1987
CA All permits requiring mitigation, 1971- 324 368 387
November 1987 ' "
An Approach to Improving Decision Making in Wetland Restoration and Creation
12
-------
tered into a database, and periodically analyzed and reported to identify trends
in decision making and areas at risk. The subset should include information
such as the specific location of the impacted wetlands and mitigation projects,
dates that permits were issued and mitigation projects were begun and com-
pleted, wetland types (e.g., according to Cowardin el: al: 1979) and areas,
functions of the impacted wetlands, objectives of the projects, arid summaries
of monitoring information. Table 2-2 lists the minimum information that we
recommend be compiled for adequate descriptions of trends in permitting ac-
tivity. Similar information also can be collected for projects implemented out-
side the permitting process. Ideally, the information compiled and reported
would be standardized nationally to facilitate comparisons between states and
regions. We recommend statewide standardization as a minimum goal.
The most accurate and comprehensive trends can be identified by compil-
ing information from the historic record, entering it into a database, and ana-
lyzing and reporting the results. For example, you could enter all Section 404
permits that required compensatory mitigation in a state into a database to
track the effects on the status of the wetland resource in the state. Examination
of the historic record can be used to: 1) identify locations with the most in-
tense project activity; 2) identify the wetland types most frequently impacted
and used as compensatory mitigation; 3) ascertain trends over time; and 4) se-
lect areas for further study. •
• Compiling the historic record, however, can be extremely time-consuming
and costly. As an alternative, we recommend that you start with the present
and continue compiling information into the future. Although this approach
will not provide you with information-on historic trends, you will have a
method of quantifying project data and detecting trends In the future. You can
always compile the historic information, as resources allow, beginning with
the present and working backward a year at a time. ;
It is essential that the minimum information recommended in Table 2-2 be
recorded for all projects and be available to resource managers. Trends in de-
cision making involving wetland projects and .their effects on the resource can-
not be evaluated unless information is compiled, entered into a computerized
database, the data analyzed, and cumulative impacts of individual projects on
the wetlands resource "assessed. See the following section, "Features of EPA's
Permit Tracking System", for insights on data quality assurance, data analysis,
and data retrieval.
FEATURES OF EPA's PERMIT TRACKING SYSTEM
An example of a data management system developed to simplify the
prdcess of entering and analyzing the information from permit records is the
PermitTracking System (PTS) (Holland and Kentula 1991). The PTS is a user-
friendly, PC based program, designed to track information ?rom three types of
" "Chapter 2: Using Existing Information
13
-------
Table 2-2. Minimum categories of data on impacted and compensatory (created, enhanced,
preserved, or restored) wetlands recommended for inclusion in a database and data
categories found in EPA's Permit Tracking System (PTS) (Holland and Kentula 1991).
MINIMUM
PTS
IMPACTED AND COMPENSATORY WETLAND
Location State and county
Specific location
Waterbody/river basin
Land use
Dates Permit Issued
Construction began/completed
Cowardin wetland types
Area of the wetlands
Contact
Location State and county
Specific location
Waterbody/river basin
Land use
USGS map name and scale
Latitude/Longitude
Township/Range/Section
Dates Permit l&sued
Construction began/completed
Cowardin wetland types
Area of the wetlands
Contact
Documents available
Reports _^
IMPACTED WETLAND ONLY
Project type
Functions documented
Endangered species names
Project type
Functions documented
Endangered species names
COMPENSATORY WETLAND ONLY
Compensation type
On-site or off-site?
Objectives stated
Endangered species names
Monitoring information
bo "as-built" plans exist?
Regular or irregular checks made?
Items monitored
Was mitigation bank used?
Name of mitigation bank
Money or land?
Compensation type
On-site or off site?
Were corrections made?
Objectives stated
Endangered species names
Monitoring Information
Do "as-built" plans exist?
Regular or irregular checks made?
Items monitored
Methods of construction
An Approach to Improving Decision Making in Wetland Restoration and Creation
14
-------
wetland permit systems: Section 404, Section 401, and state. The program In-
cludes an option to track data from other permit systems or Wetland projects.
We present the PTS as an example of a system that could be used to com-
pile, and analyze information from project files (Table 2-2). It is composed ol
two main components, data entry and query. The PTS simplifies the process o1
data entry, because in most cases, the user is merely required to check ofi
items, as opposed to entering verbiage. Standardized categories, with defini-
tions, are given for items such as wetland type, project type, and. wetland func-
tion. Selecting items and entering minimal verbiage eliminates most of the er-
rors typically associated with data entry. The PTS also sorts and prints all the
items listed in each category, making it easy to recognize information that has
been entered incorrectly. For example, if a list of county names includec
CENTER and CENTRE, it would be simple to recognize the error in data entry,
After data have been entered, corrections, deletions, and additions can be in-
corporated into the database.
The menu-driven query component of the PTS allows the user to generate
queries using the contents of the database (Figure 2-1). The program identifies
all possible combinations of queries and compiles the answers, which can be
viewed on the screen, copied to disk for conversion to tables and figures, 01
printed as hard copy. -..,•••.
The PTS not only eliminates the potential errors inherent to querying ir
other software packages, but also substantially reduces the time required foi
analyses. For example, analysis of the Oregon database using dBase III+ took
approximately three weeks. When we tested the PTS by reanalyzing the Ore-
gon data, analysis time was reduced to only three days. Furthermore^analyses
using the PTS Involve minimal user time. For example, using the PTS to calcu-
late the number of impacted wetlands and mitigation projects for each wetlanc
type entails setting up only one query, which takes approximately two min-
utes. The computer can then be left unattended as the PTS calculates the re-
sults. Traditional software packages require the user to enter a query for each
wetland type, entailing substantially more user time.
Although it was designed to track information from the permit record, the
PTS, or a similar data management system, could be used to track projects im-
plemented outside the permitting process. The key concept is that informatior
from project files must be compiled~and entered into a database, so the data
can be analyzed and made accessible to those making resource management
decisions. :
INCORPORATING ADDITIONAL INFORMATION ..j ..:. :
The process of compiling information does not stop when the details of the
project plan or permit are documented. All the project information that is col-
lected should be incorporated into the file so-that the record1 is as complete as
Chapter 2: Using Existing Informatior
15
-------
1. QUERY: Was there a net change in wetland area as a
result of Section 404 permitting?
2. QUERY SCREEN FROM PTS:
CASE
AREA L
AREA H
.T.
.F.
.F.
.F.
.F.
.F.
Impacted
Created
Enhanced
Preserved
Restored
'^X*' ,VH|
1^5:1®
£§5f
&"$&&'
•S^j^
S:SS
o.o .
9999.9
A "T" (True) in the top row
indicates that information is
needed for all subsequent rows.
The range columns define
the upper and lower limits
for area.
3. RESULTS SCREEN FROM PTS:
CASE AREA (acres)
4. RESULTS:
Area compensated:
Area impacted:
Impacted
Created
Enhanced
Preserved
Restored
<;?x;-
^\ ''•?*;
" - '>, ",','•
'^ *T
,. **vJ
J> V<
^'?5*i
^i
X '/*«•%
''*<**%
^5
90.0
12.0
4.0
10.0
24.0
12.0 + 4.0 + 10.0 + 24.0 = 50.0 acres
-90.0 acres
I Net change in area
-40.0 acres
figure 2-1. Examples of the query and results screens from the Permit Tracking System (PTS)
(Holland and Kentula 1991) generated to answer the question: "Was there a net
change in wetland area as a result of management decisions?"
An Approach to Improving Decision Making in Wetland Restoration and Creation
16
-------
possible. A seemingly unimportant fact can become a key piece of data. For
example, it is important to include monitoring data in the project file. You car
then use it to determine if the project is in compliance with the permit specifi-
cations and project objectives. The monitoring information from all similai
projects can be entered into a database and analyzed to determine if the pro-
jects are functioning as planned and if management objectives are being mel
(see Chapters 4, 5, and 6). Information that is related to the overall manage-
ment strategy should also be entered into a database, analyzed, and the results
used to make decisions and evaluate the strategy. If all pertinent project infor-
mation is available for use in decision making, management strategies and reg-
ulatory decisions will be based on the most up-to-date, scientifically defensible
information.
REPORTING THE INFORMATION
: Regular reports summarizing information in the project files are necessary
to provide a comprehensive assessment of the Wetland resource for areas of in-
terest, such as states, regions, watersheds, or ecoregions.. Furthermore, report-
ing provides a mechanism for assessing risks to wetlands. For example, trends,
such as the loss of certain wetland types, can be identified with regular report-
ing. Once trends are identified, actions can be taken to aypid losses in:wet-
land numbers, types, functions, and area. Dissemination of the reports tc
local, regional, and national authorities is critical if information in the reports
is to be reflected in management decisions. Finally, regular; reports will pro-
vide a mechanism for policy makers and planners to receive the information ir
a usable format. . ; .. ..,.._ .., .;;;__..,.•„;••-...••
Our analyses of Section 404^permitting'.during the1970s and 1980s re-
vealed several notable trends that could be used by resource managers in eval-
uating the effects of permit decisions. In most of the states studied, more wet-
land area was destroyed than was required to be created or restored, resulting
in net losses in wetland area. Furthermore,'I ess than 55% of the permits for all
eight states analyzed required that the mitigation projects be monitored by al
least one site visit; the range was from 0 monitored in Arkansas to 52% moni-
tored in Texas (Sifneos et al. In press(b)) (Figure 2-2). The wetland types of the
mitigation'projects often differed from the wetlands destroyed, resulting in net
looses in area for certain wetland typesT For example, as a result of permits re-
quiring compensatory mitigation, palustrine forested wetland was the wetlanc
type subject to the greatest loss in area in California (-143-9 ha) (Holland anc
Kentula In press) and Louisiana (-414.3 ha) (Sifneos et al. In press(a)), whereas
palustrine emergent wetland was the type subject to the greatest loss in area ir
Oregon (-15.0 ha) and Washington (-9.7 ha) (Kentula et al. 1992) (Figure 2-3).
Permitting activity involving compensatory mitigation was concentrated near
urban areas in several states*-' In-Oregon, it occurred near Portland and Coos
, , „.,_-. Chapter 2: Using Existing fnformatior
17
-------
100
0)
C
c
o
en
BC
'iZ
'-}
cr
CD
V)
CD
Q.
90
80
70
60
50
40
30
20
10'
OR
WA
CA
TX AR
State
LA
MS
AL
Figure 2-2. Comparison by state of the percent of the Section 404 permits requiring compen-
satory mitigation that specified monitoring the project with at least one site visit.
n _ , | h
• Palustririe Forested Wetlands j
CO
.g -200
S-30Q
co
CD
03
I1
u
Palustrine Emergent'Wetlands [
Figure 2-3. Comparison by state of the net change in area of palustrine forested wetlands and
palustrine emergent wetlands involved in Section 404 permits requiring compensa-
tory mitigation over the time period analyzed (see Table 2-1). The data were
obtained from the Section 404 permit record.
An Approach to Improving Decision Making in Wetland Restoration and Creation
18
-------
Bay, and in Texas it was clustered around the Dallas-Fort Worth metropolitan
area. In addition, Section 404 permitting destroyed endangered species habi-
tat in most states evaluated. The trends in: permitting described above were
obtained by compiling and analyzing portions of the Section 404 permit
records. However, unless such trends are reported, the effects of management
decisions on the wetland resource will remain unknown.
SUMMARY
The information in wetland project files must be updated, compiled, ana-
lyzed, and reported, if the data are to be of use in protecting the wetland re-
source. Trends and patterns in the information can be used to identify impor-
tant issues for further examination, which; in turn, can guide management
decisions. For reliable descriptions of the effects of management decisions,
complete and accurate information is required for all projects. Only with im-
proved documentation and regular reporting can we credibly assess the cumu-
lative effects of decisions involving individual or small groups of wetlands on
the resource.
Chapter 2: Using Existing Information
19
-------
.ill'' I"!' II
(S)
(4)
(5)
USING EXISTING INFORMATION (2)
MONITORING (4)
EVALUATING THE DATA (5)
DEVELOPING PERFORMANCE
CRITERIA (5)
(6)
(6)
IMPROVING DESIGN
GUIDELINES (6)
INFLUENCING FUTURE
DECISION - MAKING
-------
CHAPTER 3
Setting Priorities and
Selecting Sites
The ability to implement the WRP Approach successfully hinges on good
planning. Often the resources of an agency or organization are limited, so we
recommend that priorities be set before instituting a monitoring program. In
this chapter we discuss how to target sampling to critical areas and wetland
types, and present a procedure for selecting both natural wetlands and projects
to monitor. We use our .experiences studying mitigation projects to illustrate
the process of setting priorities and selecting sites. However, note that the ap-
plication of the process described is not limited to mitigation projects, and that
we present but one way that a statistical population of wetlands can be de-
fined.
To simplify the presentation, this section is written as though the Ap-
proach will be implemented in one area and with one group of projects (e.g.,
one wetland type and size class). In fact, the procedures can be used to iden-
tify more than one area or group of projects.to be monitored. The scope of the
application of the Approach is determined by the needs of the agency or orga-
nization, and, ultimately, the status of the wetland resource. ;
DECIDING ON A SAMPLING STRATEGY
. The hypothetical performance curve described in Chapter 1 displays the
changes in function over time in wetland projects as compared to similar nat-
ural wetlands. Wetland function is typically measured using an indicator. The
curve can be generated in two ways depending on how you sample. One
method is to follow the development of similar aged projects by repeatedly
" " ~" -- Chapter 3: Setting Priorities and Selecting Sites
23
-------
I I
sampling the same projects and natural wetlands over time (Figure 3-1 a). The
other is to gather data from projects and natural wetlands at one time, docu-
menting project development by sampling projects representing a range of
ages (Figure 3-1 b). The latter is how we have generated performance curves to
date. , . ,
The sampling approach chosen will depend on the history of wetland
restoration and creation in your area. If you are just beginning to construct
wetland projects, or if a number of projects are constructed every year or two,
you may want to use the first approach and follow groups through time. If
wetland restoration and creation has been going on for some time, you may
want to use the second approach and sample projects that represent a range of
ages. This approach has the advantage of potentially generating a large por-
tion of the performance curve at once. However, the design of the oldest pro-
jects may be quite different from that of the newest projects. The effects of the
different designs may confound the results so that the pattern of project devel-
opment is not apparent.
IDENTIFYING PRIORITY AREAS
We recommend that monitoring efforts be targeted to areas at greatest risk.
Areas at risk are those where the greatest wetland losses in terms of area, eco-
logical function, and/or value have occurred, are occurring, or are anticipated
to occur. In this case, value represents the benefits of the wetland that are re-
alized or recognized by society and includes uniqueness and rarity (Leibowitz
et al. 1992). In addition, the areas should have a high probability of producing
useful information. However, in some cases an area with a low probability of
producing useful information will be favored because the area or the wetland
type is so important that any information obtained will be of great benefit. De-
pending on the causes of the losses, the areas chosen will probably be places
where there is an abundance of Section 404 and associated permitting activity.
Consequently, these areas will be places where many wetland restoration and
creation projects are occurring or will be occurring.
One way to identify the areas at risk is to survey the personnel involved in
wetland management and, in particular, permitting. Unfortunately, because of
the high turnover in regulatory personnel associated with permitting, this ap-
proach will not always provide the best answers. In the short term, however, a
decision based on the "institutional memory" will probably identify current
problems and allow implementation of a monitoring program.
Examination of actual records of wetland losses, restoration efforts, mitiga-
tion projects, and growth and development is probably the best way to select
areas at risk. Chapter 2 describes a system that can be used to compile infor-
mation for this purpose. However, if the information is not available in an eas-
ily retrievable form, it will probably take a major effort to compile materials
An Approach to Improving Decision Making in Wetland Restoration and Creation
24
-------
m uoiiouoj. }o ajnsB9|/\|
UOJJ.OUI1J J.O 9JHSE81AJ
en
eo
-— Chapter 3: Setting Priorities and Selecting Sites
25
-------
from files into a computerized database. Therefore, the time and resources
needed to collect and organize the information must be taken into account in
planning. For instance, it may be more expedient to concentrate on a port.on
of the record recommended by local staff rather than delay implementing the
monitoring program while the entire record is compiled and analyzed. The
FWS's National Wetlands Inventory (NWI) reports (e.g., Prayer et al. 1989) are
also excellent sources of information on trends in wetland area for the locales
for which they have been produced.
The permit database compiled for Oregon (Abbruzzese et al. 1988, Kentu-
la et a! 1992) and similar information found in Florida (Brown 1991) were
used to identify mitigation projects for two of our field studies. For example,
the record of permits issued in Oregon from January 1977 through January
1987 indicated that 31% of the permits requiring compensatory mitigation in-
volved wetlands in the Portland Metropolitan Area (Kentula et al. 1992) (Figure
3-2) Because Portland continues to be a major growth area, pressures to de-
velop wetlands will probably escalate. Therefore, the Portland Metropol.tan
Area is considered an area at risk and a priority for management. This cluster-
ing of permit activity is not unusual. Figure 3-3 illustrates the patterns we
identified in California (Holland and Kentula In press) and Louisiana (S.fneos et
al. In press(a)).
SELECTING SITES
Once the area at risk is identified, the next steps are to define the appropri-
ate populations of wetlands to sample and to select a representative sample of
sites from each. In the following sections we will describe how to: 1) define
the population of projects to be sampled; 2) use information on project loca-
tion to define the boundaries of a study area; 3) define the population of natur-
al wetlands to be sampled in terms of the characteristics of the population of
projects; and 4) finalize the list of projects and natural wetlands to be sampled.
Defining the Population of Wetland Projects to Sample
How you define the population of projects to be monitored will influence
the definition of the population of natural wetlands to-be sampled as well as
the choice of measurements, the timing of sampling, and virtually every aspect
of the monitoring scheme. As a rule, document the discussions and decisions
that occur during planning. In particular, you should record what was consid-
ered in defining the populations of projects and natural wetlands and the out-
come of these decisions. The data used in making the decisions should also
be included in the documentation. Sometimes the decision steps that lead to
an approach or justify a choice of measures are forgotten over the course of a
study. Often such information is key to guiding the analysis of data and inter-
pretation of the results.
An Approach to Improving Decision Making in Wetland Restoration and Creation
26
-------
3
DO
-ST
Chapter 3: Setting Priorities arid Selecting Sites
27
-------
Each symbol roprosortts one or more activity
iciatod with a 4O4 permit Issued
Impacted wetlands
O Impacted/Created watiandi T
A. Created wetlands „
Figure 3-3 Patterns of Section 404 permitting in California and Louisiana, a) The number of
permits requiring compensatory mitigation from 1971 -1987 is illustrated by county
for California. Other patterns in permitting in California are discussed in Holland
and Kentula (In press), b) Locations of permitted activity involving freshwater wet-
lands from January 1982 - August 1987 are illustrated for Louisiana. Other patterns
in permitting in Louisiana are discussed in Sifneos et al. (In press(a)).
An Approach to Improving Decision Making in Wetland Restoration and Creation
28
-------
Start by obtaining a list of all the projects located in the area at risk. For
example, you could obtain a list of all the mitigation projects in the area, or
access the files from which such a list can be generated from state permitting
agencies or the U.S. Army Corps of Engineers' (COE) District Offices. Also, the
Soil Conservation Service should be able to provide information on restora-
tions done under the 1990 Farm Bill (Food, Agriculture Conservation and
Trade Act of 1990 (P.L. 101 -624)); the FWS, on restorations done under the
Waterfowl Reserve Program.
Before spending any time "digging" through files, take some time to define
the types of information that would be helpful in organizing .sites into mean-
ingful groups. This increases the probability that you will find the information
you need to make a decision as to whether a site is part of the population of
interest. Also, knowing what you want will reduce the number of trips to the
files. Figure 3-4 is a form we have used for compiling information we used in
site selection. At minimum you will need to know the location of the project,
the wetland type and size, and the property owner or a contact for the project.
Because much of our Approach focuses on the development of projects
over time, it is advantageous to know when construction was completed to de-
termine the age of the project. However, we have found that'this information
is often hard to obtain. Knowing the actual date a project was completed
would be ideal, but the, general time of year is also helpful, as is the year of
construction (e.g., fall 1987).
Examine the list of projects to identify populations of projects. For exam-
ple, Table 3-1 lists the freshwater mitigation projects in Portland, Oregon, re-
quired in permits issued by the COE and the Oregon Division of State Lands
from January 1987 through January 1991. From this information we decided
to concentrate on the most frequently occurring type of project—a wetland
that was primarily emergent marsh and open water (Figure 3-5). This gave us
a reasonably large pool of sites that represented a range of sizes and ages and
defined the first characteristic of the population to be sampled—wetland type.
Depending on the size of the population, either the entire group or a ran-
dom sample of the sites can be used. As you collect data, the variability in the
measurementsjaken will indicate whether thejiumber of wetlands sampled is
adequate. If th'e variability is large, the sample size may^need;to be increased
to improve the precision of your estimates. On the other hand, if the sites are
homogeneous, you may be able-to decrease the number of sites sampled and
save resources. Although it is best to have a large number of projects to
choose from, the number of sites available,should not be a constraint in imple-
menting the Approach. For example, we detected statistically significant dif-
ferences between created and natural wetlands for certain variables with a
sample of nine created and nine natural wetlands in the Florida Study, and
with 11 created and 12~natural wetlands in the Oregon Study.
— :—!—"—: r~~~^—^^ ~ '. ~ " ~- ~. 'HTZ" ITT ~~- ~~£ ;"'"-"" ' ' ~
Chapter 3: Setting Priorities and Selecting Sites
29
-------
COE permit number_
Stale permit number_
Date permit issued _
Mitigation type-Select [1]
O Created O Enhanced O Preserved O Restored
Permit Tracking System
COMPENSATORY WETLAND DATA FORM
Form designed by C.C. Holland and R.Q. Gibson
ManTech Environmental Technology, Inc.
U.S. Environmental Protection Agency,
Environmental Research Laboratory
• 200 SVV 35th Street ' .
Corvall'is, OR 97333
State
State.
County_
. County_
Township & Range_
Latltude/Longltude_
TOTAL]
_Sectlon(s).
Was the mitigation project Off-site or On-site?
Documents available-
Select [0-4]
O Maps
O Blueprints
O Ground photos
O Aerial photos
USGS/NWI map name_
Scale 1:.
Select [1]
O Water Body
O River Body
Water/river body name.
Specific location.
Date construction began __/__/
Date construction completed / /
Were mid-course corrections made? Yes / No
(Make notes in comments section)
ACRES
ESTUARINE
O subtidal aquatic bed ." .'•
O sublldal open water
O subtldalresl '•.''• "
O sublldal rock bottom
O subtldal unconsolldated bottom.
O Interfidal aquatic bed
O Intertidal baach/bar : '"..'
O Intertidal emergent
O iitertdal Bat '-.:"''•
O kitartdal forested
O Intertidal reel..
O Intertidal rocky shore
O talertldal scrubfehrub
O Wertldalstreambed
O inlertdal unconsoMaled shore
LACUSTRINE
O Smnotlc aquatic bod
O Imnatfc open water
O limnetic rock bottom
O (mnoUcunconsoBdated bottom
O littoral aquatic bed
O taoral beach/bar
O littoral emergent
O littoral flat
O littoral open water
O littoral rack bottom
O littoral rocky shore
O littoral unconsolidated bottom
O littoral unconsolidated shore
RIVERINE !
O tidal aquatic bet)
O tidal beach/bar I
O tidal emergent
O tidal Bat
, O tidal open water,
O tidal rock bottom
O tidal rocky shore
O tidal streambed|
O tidal unconsolidated bottom
O tidal unconsolidated shore
O tower perennial laquatlc bed
O tower perennial beach/bar
O knver perennlaljemergent
O lower perennial .flat
O lowsrperenntal open Water
O lower perennial rock bottom
O lower perennial rocky shore
O tower perennlal'streambed
O lower perennial unconsolidatad bottom
O lower perennlaljunconsolldated shore
O upper perennial, aquatic bed
O upper perennial beach/bar
O upper perennial fiat
O upper perennial open water
O upperperenniafrock'ooltom
O upper perennial rocky shore
O upper perennial streambed
O upper perennial unconsolidated bottom
O upper perennial unconsolidated shore
O intermittent aquatic bed
O intermittent beach/bar
O intermittent flat;
O Intermittent open water
O intermittent rock bottom
O Intermittent rocky shore
O intermittent streambed
O intermittent unconsolidated bottom
RIVERINE (cont)
O unknown perenhlal aquatic bed —•—
O unknown perennial beach/bar .—
O Unknown perennial flat •—
O unknown perennial open water —.—
. o unknown perennial rock bottom —.—
O unknown perennial rocky shore —•—
^ O UnknownVerertnlal streambed —•—
O unknown perennial unconsolidated bottom —•—
, O unlinown perennial unconsoiidated shore —,—
PALUSTRINE
t O aquatic bed —•—
O emergent —•—
, o Hat —'—
O forested —-—
. O moss/lichen '—
O open water —•—
. o rock bottorri —'—•—
O scrub/shrub —•—
^ o uncortsolldated bottom —•—
O unconsolldated shore —•—
MARINE
~ o subtidal aquatic bed —•_
O subtidal open water —•_
; '0 subtidal reef ' _-i_
O subtidal rock bottom —.—
' o subtldal Unconsolldated bottom —._
Olntartldal aquatic bed —•—
' o Mterfidal beach/bar - _•_
O Intertidal flat —•_
O hterlidalreef —•_
O Intertidal rocky shore —•_
O Intertidal unoonsolldated shore _^-_
TOTAL AREA
! ' . , "i I , . 'I
Figure 3-4. An example of a form that can be used to compile information on wetland projects
(Holland and Kentula 1990)]
An Approach to Improving Decision Making in Wetland Restoration and Creation
30
-------
Titls
Author's First Initial Middle Initial Last Name_
Year ' Source
Content.
First Initial
Organlzation_
Address
City
Middle Initial Last Name_
State
Zip
Phone ( ) .
VSS*^ ^jS^" '•£>'
. '. -.'•• •> * •&.•?<•.«.' '
First Initial
Organization..
Address
City
Middle Initial Last Name_
State 2ip_
Phone ( " ) ,
^^i^
First Initial
Organization..
Address
City
Middle lnitial_ _ Last Name
State
Zip_
Phone ( ) .
'9S-£M, yt^^gl'^^MflgRy'X^I::^ ^V«i/^-4:.>i *%&1&!&£>&*?&&'.
First Initial
Organization..
Address
City_
Middle Initial Last Name_
State
Phone ( ')
Obieclive:
Method:
As-built:
Figure 3-4. (continued)
Chapter 3: Setting Priorities and Selecting Sites
31
-------
able 3-1 Numbers of freshwater mitigation projects in Portland, Oregon, by wetland type and
size required in Section 404 permits issued by the U.S. Army Corps of Engineers and
the Oregon Division of State Lands from January 1987 through January 1991^
"YPE
0-2
SIZE (Acres) TOTALS
2-4 4-6 6-8 8-10 >10
/larch
'ond
(tarsh and Pond
•tosh and Shrub-scrub
vlarsh and Forested
rfarsh. Shrub-scrub, and Forested
vlarsh. Pond, Shrub-scrub, and Forested
vlarsh. Pond, Aquatic Bed, and Forested
vJarsh, Pond, and Flooded Grassland
'ond, and Riparian
'ond, Forested, and Stream Channel
iiverine Wetland
Stream Channel
Creek Bank
1 00-year Floodplain
Mud Flat
Unknown
TOTALS
9
24
4
2
0
1
0
0
0
10
1
0
1
1
1
1
3
58
5
0
1
2
0
0
1
1
0
0
0
1
.0
0
0
0
0
11
o
1
0
1
0
0
0
0
1
0
0
0
0
0
. o.
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
...„„,..
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
2
0
1
0
0
0
0
1
0
0
0
• 0
0
0
0
0
0
0
2
2
12
3
1
1
0
0
0
0
0
0
0
2
0
0
0
10
31
16
39
8
6
1
1
1
10
1
1
3
1
1
1
13
107
Figure 3-5. Example of a typical mitigation project sampled in the Oregon Study,
An Approach to Improving Decision Making in Wetland Restoration and Creation
.32
-------
If the number of projects being considered is large, you may want to seel
permission for access at this point to prevent wasting time gathering informa
tion on sites that are not accessible to the study. See the section on finalizing
the lists of sites to be sampled for ideas on how to obtain permission for ac
cess.
Defining the Boundaries of a Study Area
.In the process of setting boundaries for the study area, the definition of th«
population of projects is refined to include the concept of .ecological setting
The study area can be defined as either the entire area at risk or just a portior
of it. The study area should bound a population of projects in as homoge
neous an ecological setting as possible. By this we mean that the boundarie:
should be set to include similar hydrologic, climatic, geologic, or other rele
vant geographic conditions that influence the nature of the wetlands.
Taking a regional perspective
The boundaries of the study area are set by the following procedure adapt
ed from Abbruzzese et al. (1988). The first step is to examine the distributior
of the projects relative to the ecoregion boundaries (e.g., Omernik 1987) to de
termine which ecoregions to include in the study area. We recommend tha
sites within the same ecoregion be considered a population. .If the area at risl
is large, it is advisable first to plot the. locations of the projects'and the ecore
gion boundaries on 1:500,000 scale state maps. Then the overall pattern car
be analyzed to identify smaller areas on which to concentrate.. If you decide
to subdivide the area at risk, consider,only those smaller areas in subsequen
decisions. The next step is to transfer the locations of the projects and ecore
gional boundaries to 1:100,000 scale NWI maps for comparison with U.S. Ge
ological Survey (USGS) topographic or ecoregion maps at the same scale. Ex
amine spatial patterns of relief, hydrographic features, and vegetative cover tc
identify possible subregions or discontinuities. For example, in the Oregor
Study our examination of the topography around the Portland Metropolitan
Area showed three distinct subregions (Figure 3-6). The Coast Range extendec
into-the northwest quarter ofthe city. Small hills dominated the south. Low^
lands created by the Columbia and Willamette Rivers occupied the north anc
west. Most of the mitigation projects were located in the lowlands, therefore
the lowlands within the Portland Metropolitan Area defined the boundaries o
our study area, and only projects within that area were part of the populatior
studied. .:.-:.. . . .; ----- ••
Considering ecological setting
• Matching the ecological setting of natural wetjands with that ofthe pro-
jects is a fundamental aspect ofthe Approach. O.ur initial tendency was to se-
Chapter 3: Setting Priorities and Selecting Sitet
... . 33
-------
O
£
•o
>*§
& 2
0
v» o>
O C
O O
OK
\
, 13
S J
rt 5
— -E
o ^
« =:
S I
O «
O~ >•
I
« 'E
.!-§
JD rt.
14
'C
^ T2
-------
lectthe most pristine sites in the ecoregion as the natural wetlands to be used
for comparison. For example, Brooks and Hughes (1988) recommended that
reference sites should be relatively undisturbed and representative of the re-
gion and the population of mitigation sites. However, examination of Section
404 permits requiring wetland creation in Oregon showed the majority of the
created wetlands were in or near metropolitan areas (Kentula et at. 1992). Be-
cause this pattern also occurred in several of the other states we studied, ques-
tions were raised about whether relatively pristine or undisturbed wetlands
were legitimate comparisons for projects located 'in a human influenced set-
ting. We also realize that these questions are related to defining attainable and
acceptable performance for the projects. Natural sites chosen from landscapes
with limited human influence may not reflect the potential structure and func-
tion of projects found in an urban setting. In other words, comparison of wet-
land projects with natural wetlands located in a similar land use setting and,
therefore, exposed to similar ecological conditions, is needed to ensure that
what is "expected" of a project is within the bounds of possible performance
given the setting.
Land use has become a major part of our definition of ecological setting.
For instance, the Oregon Study involved projects and natural wetlands within
the Portland Metropolitan Area. While we felt that it was legitimate to com-
pare projects located in an urban setting with natural wetlands in the same set-
ting, we realized:that wetlands, in urban areas may not be of the same quality
as wetlands in other land use settings. We knew it would be valuable to docu-
ment those differences and use the information to direct wetland protection
and .restoration to areas with the greatest potential for ecological benefit. In
the case of mitigation, such information can be used to avoid compensating
for wetland losses with projects of limited ecological value due to their loca-
tion. In addition, we realized that the information could be used to identify
how particular land uses impact wetlands and how to buffer the systems from
these impacts. The nature of the interactions among various anthropogenic
factors and between anthropogenic arid natural variables is a legitimate eco-
logical research topic and one of increasing importance. Knowledge of the
relative .influences, of urban and natural environmental forces on ecosystem
function is fundamental to our understanding of ecosystems and the impacts of
human activities on them. The necessity of such information was acknowl-
edged in a special feature on urban gradients published in a recent issue of
Ecology (Volume 17, Number 4, 1990).
We recommend that the populations of wetland projects and natural wet-
lands be stratified by land use setting so that the natural wetlands represent the
various land uses surrounding the wetland projects. In the Florida Study, we
. used the Landscape Development Intensity (LDI) index to quantify the devel-
opment intensity surrounding the wetlands and to stratify the sample (Brown
1991).
Chapter 3: Setting Priorities and Selecting Sites
35
-------
Defining and Sampling the Population of Natural Wetlands
In the Oregon Study the population of natural wetlands sampled was de-
fined in terms of the population of projects, (i.e., palustrine emergent marshes
with.open water that were less than or equal to one hectare in size and located
in the Portland Metropolitan Area within the lowlands created by the Colum-
bia and Willamette Rivers (see Figure 3-5). We randomly selected natural wet-
lands from the population defined using a procedure adapted from Abbruzzese
et al. (1988). An overview of the procedure is given below, for additional de-
tails see Abbruzzese et al. (1988) and Brown (1991). Certainly this and other
methodologies can be adapted to meet your individual needs.
In our procedure, first we overlay a grid, with each cell representing 260
ha, on a 1:100,000 scale USGS topographic map marked with the study
boundaries. After sequentially numbering the cells falling within the study
area, we transfer the numbered grid to 1:100,000 scale NWI maps. The order
in which to sample the numbered cells is determined randomly (e.g., with a
random number table). We then identify and number sequentially all wet-
lands meeting the specified criteria (e.g., wetland type and size) in each cell
sampled. The NWI codes on the map are used to identify wetland type. Wet-
land size can be measured using a template. For example, we used a transpar-
ent grid with 64 cells, with each cell equal to approximately 4 ha. Therefore,
a wetland that filled one-fourth of a cell would be 1 ha in size.
The number of grid cells to be sampled was determined by calculating a
progressive mean (Marsh 1978). We sampled five cells at a time from the ran-
domly numbered list and calculated the mean number of wetlands per cell
meeting our criteria. We repeated this procedure until the mean number of
wetlands per cell did not change more than 0.1 and three times the number of
wetlands desired were identified. These numbers were chosen to increase the
probability that we could obtain a representative sample of the wetlands in the
area and have a large enough number of potential sites so that sites eliminated
(e.g., because access was denied) could be replaced.
We performed a 5% quality control check on the wetland area measure-
ments taken from the NWI maps, i.e., a second individual repeated 5% of the
area measurements, so that we could assess measurement precision. An error
level of less than 5% was achieved in our studies. We suggest that the project
leader and statistician define meaningful values for the quality control check
and the acceptable error level for your particular study.
Finalizing the List of Projects and Natural Wetlands to be Sampled
The activities described above will result in lists of projects and natural
wetlands that have, thus far, met the sampling criteria. Replace any sites elim-
inated from consideration with the next wetland on the appropriate list. Sites
are eliminated if: 1) access is denied by the landowner; 2) the wetland project
An Approach to Improving Decision Making in Wetland Restoration and Creation
36
-------
has not been constructed or the natural wetland has been destroyed; 3) a field
reconnaissance of the site reveals that it does not meet the specified criteria
(e.g., it is the wrong type or size); or 4) conditions on or near the site would be
hazardous (e.g., garbage is actively being dumped on the site). We have had
to eliminate sites from our lists for all of these reasons. The following discus-
sion describes how to confirm that a site is, indeed, suitable for inclusion in
your study.
The first step, if it was not done earlier, is to obtain permission from the
landowner to enter the wetland. A contact person is often listed in the permit
or project files. Finding the owner of the natural wetlands is more problemat-
ic. You can use ad valorem tax maps of the county, visit the site and use de-
tective skills (e.g., talk to adjoining property owners), or contact the owner b>
mail and then follow up with a phone call. Be prepared for many individuals
to deny access to their property. In the Oregon Study/owners denied access
for 35% of the natural wetlands from the list of potential sites (Abbruzzese el
al. 1988).
After you have eliminated the sites to which access was denied, you musi
locate each of the remaining sites to ascertain if the wetland projects have
been completed and the natural wetlands still exist. Locating the projects anc
natural wetlands will probably be time-consuming because project files ofter
contain-vague information on the locations of the wetland iprojects, and it i;
often difficult to locate the natural wetlands from maps. While at the site doc-
ument the location, ease of access, wetland type, and surrounding land use.
Figure 3-7 is an example of a form that could be used. By following the proce-
dures'outlined above you will have finalized a list of projects and natural wet-
lands to be sampled. You also will have compiled information on each site
that will be useful as you begin monitoring.
SUMMARY
This chapter presents a strategy for setting sampling priorities and selecting
sites for a monitoring program. We recommend that priorities be based or
past and projected impacts to wetlands so that efforts will !be targeted to lo-
cales where the resource is at greatest risk. We also recommend that the eco-
logical setting, in particular the land use surrounding a wetland, be accountec
for in site selection. Only when wetland projects are compared with natura
wetlands located in a similar setting and exposed to similar ecological condi-
tions, can the performance criteria for a project be within the bounds of wha
is possible.
Chapter 3: Setting Priorities and Selecting Site.
37
-------
FORM I: GENERAL SITE INFORMATION
SITE NAME/CODE 5Ofe5 "Fosbert)" STATE,
Date
COUNTY
WETLAND TYPE PU.g
PERSONNEL NAME
A. Describe ease of access to and within the site (roads, parking, problems due to water depth,
etc.).
£asf/y accessible. In houf/'nQcle\e/oprr)ej~>-l;~-can pccrH anywhere.
B Provide directions to site. Attach a marked copy of a map if needed.
£-5 Atorfh. To«e Carmen Rd.exC+. Turn right- on Carmen Rd CQO tost-). Tu-rn left on SW
Meadowu. Tu-rr, riahf-on Hruse Wfaccfe. «ruse. uJoods turns info l*)e-sHoJi& Dr. -then
into fbsb&rafd.'ni.rn riqh-t-on Bay Crce.K Dr Wetland f-S an n'ghf ds£ ofBAy creeK/
C. Document check of ownership of site^Was the owner contacted? $e§yNo
Was trespass permission granted? ^sCNo
D. Sketch the wetland below. Include information on the factors influencing hydrology (e.g., water
control structures, ditching). Sketch in the landuses and natural cover on the wetland and in
the surrounding area (use categories listed on page 2). Indicate north.
Figure 3-7. Example of a completed form that can be used during a field reconnaissance to
collect information on potential study sites.
An Approach to Improving Decision Making in Wetland Restoration and Creation
38
-------
FORM I: GENERAL SITE INFORMATION Page 2
I. Indicate' % open water, % vegetated and % non-vegetated areas within the wetland (A-C
should add up to 100%):
A. BO % open water
1. ?5 % unvegetated
2. J % with submerged aquatic vegetation
B. f5 % vegetated
1. O % trees
2. Q % shrubs (15 feet or less)
3. 15 % herbs . . . . .
C. 5 % unvegetated
TOTAL 100%
II. ' Indicate % relative cover of surrounding areas within 100 meters of the wetland
boundaries (A-E should add up to'100%}:
A. 1° % trees
B. 2- % shrubs
C. 8 % natural herbaceous vegetation
D. O % water body-specify type: .
E. 8O % human landuse
. 1. O % crops " " - - ......
2. O % fallow
" •' :'-';:-•-.,"•.-.••• -. 3. o % grazing - •'-.' -• ~
4. O %. industrial-specify type:_._ - .
5. Q% commercial
6. 5 % transportation corridor
7. ?-5 % housing-single family dwellings _
_ ,_,„_.,., <;J - e.:'' O % housingf-rriultiplefamily dwellings , ...
TOTAL: 100% **NOTE: 1-8 should total the percentage value in E.
III. Indicate % of wetland which is disturbed and describe the disturbance (for example,
ditches, water control structures, dumping, fill, and anything that might be hazardous):
5°/o dft-rui-bed - plas-Kc and hay bales :
3 pipes
some, s-hxM&5 lef-f on sife
IV. Comments: ' . •. • - -
a5i'ng
Figure 3-7. (continued)
Chapter 3: Setting Priorities and Selecting Sites
39
-------
(5)
(4)
(5)
USING EXISTING INFORMATION (2)
SETTING PRIORITIES/
SELECTING SITES (3)
EVALUATING THE DATA (5)
DEVELOPING PERFORMANCE
CRITERIA (5)
(6)
(6)
IMPROVING DESIGN
GUIDELINES (6)
INFLUENCING FUTURE
DECISION - MAKING
-------
CHAPTER 4
Monitoring Performance
Monitoring is a key element of the WRP Approach. Post-constructio
monitoring of wetlands, however, is seldom performed (Brooks 1990, Gwii
and Kentula 1990, :Kusler and Kentula 1990b). Before proposing or initiating
monitoring program it is important to develop a plan based on project objec
tives as an integral part of the project. The plan,documents proposed assess
ment procedures, timing, and frequency as well as the person or organizatio
responsible. Monitoring decisions should be guided by the type of informa
tion that will be needed to determine if project objectives are being met. Whi
is responsible for performing and overseeing corrections? Will the data fror
monitoring be used to evaluate compliance with specific permit conditions? I
the purpose of the study to detect an improvement or decline in a wetland'
condition or functions? Will the data collected be compatible with informatio
from previous reports or similar studies (Brooks 1990)?
"In the past/assessments of wetland projects have emphasized structure
rather than "functional attributes. Although structural features, such as wate
levels, are convenient to measure, these features are most useful when relate.
to the functional capability of the wetland. Many structural measures, such a
community diversity, become indicators .of function when monitored ove
time. Therefore, it is critical that a performance evaluation of a project consic
er both functional and structural capabilities (e.g., Marble 1990).
The intensity of post-construction monitoring varies with the environmer
tal significance of the project, the compliance requirements, the age of th
project, and the probability of successfully achieving targeted wetland func
Chapter 4: Monitoring Performanc
43
-------
ons. Most wetland projects are designed to provide only a few specific func-
ons. By focusing monitoring efforts on these designated functions, the associ-
ted costs can be reduced.
In this chapter we propose some general procedures for performing assess-
lents at three levels of effort. In Chapter 5 we suggest how to analyze the
ata and use the results. The three assessment levels are: 1) documentation of
s-built conditions, 2) routine assessments, and 3) comprehensive assessments.
or each of the variables suggested, a brief rationale relating it to wetland
jhctlon is provided in Table 4-1. The variables and suggested methods for
neasuring them are presented in Table 4-2 and apply to both projects and nat-
.ral wetlands. The data collected during each level of assessment are hierar-
hical. That is, information obtained during an as-built assessment forms the
lasis for routine and comprehensive assessments that occur later. A hierarchi-
,al approach to data collection facilitates making comparative evaluations
iv|r time and among similar sites. Ultimately, this process can lead to the de-
elopment of performance criteria for future wetland projects (Chapter 5).
DOCUMENTATION OF AS-BUILT CONDITIONS
We use the term as-built conditions to refer to actual project conditions at
ne time of completion. As-built assessment refers to the data collected for
valuation of this condition.
Rationale
First, we recommend checking the wetland for compliance with the design
riteria and for agreement with permit conditions or project objectives. Sel-
lom do as-built conditions coincide with original designs. Therefore, it is es-
ential that as-built conditions be documented. At this level of assessment,
.ollect baseline data on project location, morphometry, hydrology, substrate,
,nd vegetation to compare with project objectives and construction plans, and
iocument any differences.
Differences between the design and actual construction may significantly
ffect the wetland's potential performance. Assessments of as-built conditions
lejp identify noncompliance with permit conditions or project objectives so
nat corrective steps can be taken as required. Some modifications to the origi-
ia| design are expected due to unforeseen conditions that become evident
'uring construction (Gwin and Kentula 1990). There may also have been mis-
akes. If construction according to the original design or existing conditions is
kely to limit wetland performance, then corrections should be made before
ompletion of the construction phase of the project. These changes should
len be justified and accepted before the project is officially approved by the
lermitting agency or organization responsible for the project.
n Approach to Improving Decision Making in Wetland Restoration and Creation
44
-------
-a
$'
2
-a
c
I
c
|
—
ENERAL
O
i
CD
CO
3
CO
CO
ra
gj
1
a
o.
re
CD
c
75
CO
CO
_o
ro
£
determine minimum, maximum and mean
depths and slopes from topographic profiles for
each transect (Figure 4-2)
•5
o
'•£*
«
influences hydrologic gradient, plant
establishment, animal access, characte
wetted edge
«ivnM>>r«AM* .
CO
c
CO
1
Q.
O
DC-
Chapter 4: Monitoring Performanc
45
-------
SUGGESTED USE(S)
z
o
RATIONALE/FUNCTI
UJ
m
£
^
^
8
c
'£
c
o
u
YDROLOGY
ac
establish presence and periodicity of hydrology
o c
o
c *=
S c
•§ ^
^ .
provides evidence of hydrology ii
water during sampling, boundary
CO
oT
£
J2
I
r\
Indirect indii
UBSTRATE
CO
verify construction specifications
£
g
O»
13
C
CO
influences suitability as planting
medium
CO
I -
0)
'o
CO
determine extent (1) and time of formation of
hydric soils (3), boundary delineation
indicates hydric characteristics
co
O
o
U
'5
CO
verify construction specifications, benchmark
for temporal changes
^
o
CO
•d C
C 0
CO *J2
influences suitability as planting
medium, root growth and infiltra
CO
£
•i
£2
verify construction specifications and identify
potential plant propagules
provides baseline information
p
o
c/l
!5
CO
compare to natural wetlands, document II
temporal changes
•o
c
CO
_c •—
^ c
2 ~^_
•a %
C «
indicates suitability as planting a
medium, condition of soil proces
Zedler1991)
CO
^
1
CD
'c
CO
D)
0
measure rates of sediment accretion or erosion
for comparisons to natural wetlands, document
temporal changes, document/correct erosion J
c
o
'%
£
CO -
indicates potential for sediment
removai, disturbance
CO
X
Sediment fl
VEGETATION
^-
verify permit or project planting conditions,
delineation, calculate weighted averages and
ratios (see Chapters 5 and 6)
£
CO
Q.
•o
c
defines wetland type, habitat, a
diversity
to
^
in
+-*
.2
a>
.2
a.
CO
verify project goals, benchmark for future
assessments |[
influences use as habitat
-
—
^
CN
Coverage (
evaluate planting success, suggest replanting
strategies 13) |J
^
v>
t>
o
f
O)
c
indicates effectiveness of plant!
influences project goals
co
V—
a
Survivorshi
An Approach to Improving Decision Making in Wetland Restoration and Creation
46
-------
J-
I
SUGGESTED USE(S)
2
o
o
LL
L1J
_l
2
O
- r"
DC
LLl
CQ
E
^
. ;•-
: :-
1 FAUNA :,
i
o
evaluate use by common, rare, and exo
species over time
.
to
"•*-•
3'.
CO
.c
• co
CO
CD
CO
CO
T3
-£
?3
CN
Observations (1
deterrnihe habitat potential over time •
CO
'£
CO
-C
To
"•"c
JS
0
a
CO
: f
"co
a>
G
CO
^
o
Habitat evaluati
o5
>
.0
evaluate presence and abundance data
time ' •
' c '
u
c
o
o
•*-
0
Q.
3 '
O .
OJ
O
8-
O ;.
(I>
Q.
CO
•a '•
CD
O
CO
CO
"'CO ='
"co -:
CD
C
'5.
£
CO
CO
cj
'o
CD
a.
tn
t
c
D
e
. Species or comi
1 (3)
7
_
1 WATER QUALITY
W C fl)
gl a
co x 2
« ^ 0 '
provide baseline data for specific p'rojei
(.1-), evaluate water treatment function,
variations in vegetative performance, c
with faunal use (3)
',^
'O
..§
CO
CB »
•t*s>
S5
.<-• ^
•p "^
• CO 2
CD 'CO
•& 1-
i- O
3 C '
'§ 0
.> 0 .
CO ^-
•O D
.'•5 co
.£ =5 ' .
'-
_
'".
^. •
Water samples
o
i—
<
cc
o
1 ADDITIONAL INF'
« o
3 c •
5 T)
co £
benchmark for. temporal assessments,
for office review of 'wetland and surroi
buffer-
CO
. C D
°-o
C C
= JO
l.£
v- TD
D. §
' 1 g •';
il
g g
*^ c
i s .
ro S
§3
II
-2 g
Q. 0
~n
CN
-
: —
O
U .
Photographic re
j benchmark for future comparisons
c
_g
c
CO
a.
X
CD
TJ
C
CO
.c
,g
CO
^
*'~"
CO
c
_g
it:
n
CO
CO
. 0)
">
2
d
-^t
CO
CN
T-^
OS
•p
CO
| Descriptive nar
Chapter 4: Monitoring Performana
-------
LU
>
C/3
111
S
cc
a.
S
O
0
111
Z3
O
cc
AS-BUILT
LL1
m
EC
>
I GENERAL
». Tt
use existing map or create map
with property boundaries, scale
north arrow, date, latitude and
longitude, county and state (adi
addresses in urban areas and
landmarks in rural areas)
j^
, c
g
'§
o
_i
—
*W* •"""•
(O O3
£:£;
D) .
i ™
w
o> c
*!
11
_
~5
a.
.2 2
"m — Z
c ro
5 +•*
0) 0)
1^ c
•~ '-&
"53 5
S °
C
"e
i
(A
1
CD
T3
C
CO ~-~
c r^
"to _;
% "
a CD
. CD
ra £ 'o1
c " 1 °-
^15!
75 ra a> •§
c — 3
o 03 cr D.
ISII
la S u
.3,5 g- ra
O — ^ w.
| 1 o °
ill5l
> to 5
determine jurisdictional boundai
and use basic survey technique
(Rgure 4-1 ) to create a map of
project (ha)
CO
CO
CD
measure elevation changes at
intervals along transects (see
Figure 4-2, Gwin and Kentula
1990)
CD
D.
O
CO
An Approach to Improving Decision Making in Wetland Restoration and Creation
48
-------
COMPREHENSIVE |
Ul
P
o
-•-Bp
CO
in
j
CO
E
•a
CD
§
1-
o
_o_
EC
fc
O
O-
o-
O
5
planimeter boundary of wetland
indicated on the project map and
based on jurisdictional boundary
(m/ha)
~o ^
•a c £•
c to co
planimeter boundary of wetla
indicated on the project map
based on jurisdictional bound
(m/ha)
CO
^~-
o
I --
CO
o
CO
2
•5
.1
1 .
— ,
>
o
o
° tN
.2 Q. A °-
•§ "° "!>
C ^==0.
300
o co > £
3 ^0 5
i 1 1 .5-
§ 5 r~
o °?o
T3
0) CO > ti
3 °" o —
W Jfc = CO
CO m CO *
o <3 j; _to
"D ~S
C c
III I
15=-?
3 O O _
_C D> g ^
o to > ti
3 °> 0-2
w -fc = to
to TS co .
O +-* ^1 CO
E -^42=5
CO .
CN
*~*
f
o
•o
B
i
measure inflow and outflow
discharge if present (m3/s) with
flumes or weirs
•B ,
measure inflow and outflow
discharge if present (m3/s) w
flumes or weirs
CO
,_;
• ~vt
•o
CO
0
u_
use direct observation to indicate
major pathways on map
f
_o
TJ
• —
° ra
.2 ^
V« c
> ^
is en
S &
•° 3
jr
05 Q.
•^.o
in W
3 E
O
s
use direct observation to indi
major pathways on map
CO
.- ^
^.
at
o
CO
D.
I
rr
record observations of indicators
(Federal ICWD 1989) |
0
CO
u
'•5
.£
"S-
», 0>
g CO
O ^
5 Q
M O
° 15
"o !K
k. 0>
o ~a
u o>
£ 5t
M
e .
record observations of indica
(Federal ICWD 1989)
CO
o?
_^.
to
o
•M
to
^
tJ
.i
UJ
1—
1
CO
use soil auger or dig pit to depth
of compacted soil of liner (Federal
ICWD 1989)
"o j
0."^
n\ ™
use soil auger or dig pit to d<
compacted soil or liner (Fede
ICWD 1 989)
CO
,_r .
~ .
§"
"o
CO ,
Q
"o —
o 2, £
.c -§
•§ CO
III
to -a
<2 <2 C
. g .£ to
3 •§ E •r-
Q
>
M- 0
use Munsell color chart to
determine chroma and hue o
matrix and mottles (Federal I
1989)
CO
—
Q
1"
u
'5
CO
o
tN
4
_v
•§
Chapter 4: Monitoring Performance
49
-------
COMPREHENSIVE ||
LU
g
O
ti
:s
LU
m
^£
EC
^
, -a
03
a
c
'• c
1
LU
,|
, 03
use soil texture triangle to classify
soil based on feel (Ho'rner and ';
Raedeke 1 989) or standard
methods • ' " • '
'
.- '•'•
; '
>..
^
use- soil texture triangle to "class!
soil based on feel (Homer and
•Raedeke 1989) •'•'•' • -•
_
CO
fc
^
I
'5
t
document source location and
addition of any soil amendments
(e.g., fertilizer, organic matter,
salvaged marsh surface)
., ',
_'_ .
1 03"
: O
.3'
0 •:
. CO-
• 'o :
to
03
.2,^.
•0 < *-
'I II
'- 1 It
"O 4=; 0)
•*-*, v-'
C1 CD
sample during as-built assessme
. if salvaged marsh surface or oth
organic materials are added
CO
_*
w
o
£;
tO'
£•
o
'c
m :
CO
o'
install clay pads at substrate
surface as reference points
(Cahoon and Turner 1 989)
install clay pads at substrate
surface as reference points
(Cahoon and Turner 1 989)
CO
"— •
• X
«t-
c
:|
T3
03
~z.
0
1
LU
O
111
identify species and wetland II
indicator and native/introduced
status (Reed 1988)
•T3
C3
•o o
C 3
ta T3
1 t:
"O 03 —
S:> oo
•^3 00
<0 !5 03
03 '- r-
'i i -g
& (0 Q)
>. o — r
fl 5 '
o> TJ ro
S .?:• M
identify species and wetland
indicator and native/introduced
status (Reed 1 988), document
planting locations and methods
B
fc
t-~
eo
4-1
.2
8
o
o.
CO
estimate cover visually to nearest
1 0%, map plant communities, .
collect plot data along transects
(Brpwer and Zar 1 984, Leibowitz
et al. 1 991 ), collect data for
productivity studies
£ : '
CD tO
' C '.p
s! .
> E
"ra E
3 O
c
' O3 <0*
o a
° &
3 E
i ^
'« o
03 r-
+-i
Tn
estimate cover visually to neare:
10%, map plant communities
t-f
CO
CS
03
OJ
m
I
cS
visually determine % of plants
alive, tag individual shrubs and
trees |
03
_>
visually determine % of plants c
CO ••
. ' •
T—
.0.
tn '•
1
w
—
'JS
An Approach to Improving Decision Making in Wetland Restoration and Creation
50
-------
COMPREHENSIVE
LU
~2L
i
fe
g
AS-BUILT
UJ
i
m
E
>
•z.
u.
•o
c
CO
record direct and indirect
observations of wildlife, fish
invertebrates
•o
c
ra
jc
.2
£ 42
'o =5
c —
r '5
H£
CO O
Us
"O fij o
"P £ S
O ^ Q)
o J2 >
£ o.£
"-• -a
c
to
record direct and indirect
observations of wildlife, fish
invertebrates
—
CN
„
-,_
m
g
ra
<5
CO
O
in
£
"0
use Habitat Evaluation Proce
(FWS 1 980) or comparable
method for selected species
•c
.
_.'£ O.
P o>
c .c
8 I
O (O
« o
.2 *=
o '5
Q) n>
0. 0.
W CO
^
5
3
0
UJ
1
"
* T3
111
measure appropriate parame
based on project objectives '
pH, conductivity, total suspt
solids, nutrients, pollutants)
. T3
io Q «
fc- . ^
S « Sr -
measure appropriate parame
based on project objectives
pH, conductivity, total suspi
solids, nutrients, pollutants)
CO
*^L
«
"5.
w
1
g
tj
tr
o
LL
—
g
0
H-
D
Q
^
» ^
1 §
e i
photograph wetland and
surrounding landscape from
several directions with 50mi
using 35mm film from perm
photo stations (Horner and
Raedeke 1989)
2 o
§J gs .
0? .— JC
CO M Q. (U
•n ° 1 1 ""
S &e i5
•r^ m ?• ^ *n
HJ W O D. 01
5 j2 6 2 E
•aflls
ra C- O **"" J2
o 2 "g E o 5?
o *- m E **^ oo
^ 3 .t W ^5 O3
Q. « "O CO M *—
S o
Q) __ i5
> P 0
o ~ jc
CO » Q. „,
photograph wetland and
surrounding landscape from
directions with 50mm lens i
35mm film from permanent
stations (Horner and Raedek
1989)
1
u
£
.y
Q.
ra .
&". ••
*^ C»J
o
£ £ .
_•§.
describe and explain notable
features and changes for ea
major'variable
o"
j5 ra
JQ 0)
° "M
C 0)
"m OJ
— r C
0> "5 o
111
0> W *°
"'"So
JjJ ^~> .St
w S c
T3 M- E
J=
-. O
describe and explain notable
features and changes for ea
major variable
Q)
2
ra
c
_>
Q. CO
IS
I
Chapter 4: Monitoring Performana
51
-------
As-built assessments of projects or initial assessments of natural wetlands
rovide baseline information from which site development and functional per-
jrmance can be evaluated over time. For example, vegetation data collected
uring the Oregon Study could not be used to estimate the survival rate of veg-
tation planted during wetland construction because there was no documenta-
on that planting occurred (Gwin and Kentula 1990). As-built assessments
/ould have provided this information so that vegetation survival rates could
tave been determined and that aspect of the project design evaluated.
Vhat to Include
The objectives for the documentation of as-built conditions are to collect
ufficient information to assess compliance with permit conditions or project
tbjectives and to provide a baseline for future evaluation of project develop-
nent and performance. The basic elements of an as-built assessment are listed
n Tables 4-1 and 4-2. There are both graphic and written components to an
is-built assessment. Maps are generated to record wetland area, shape, the
latterns of vegetation and open water, major structural components (e.g.,
vater control structures) and surrounding land use (Figures 4-1, 4-2, 4-3). A
vritten narrative augments the graphics and serves as a record of what was
lone during construction regarding substrate, hydrology, and planting. Addi-
ional information may be required. For example, if a project objective is to
mprove or develop habitat for a specific fish or wildlife species, then an as-
HJilt assessment should include a species census, or at least an evaluation of
)otential habitat at the time of project completion.
The as-built assessment should be completed by the party responsible for
:onstruction of the project. In a regulatory situation, this can be ensured by
naking the assessment a mandatory condition of the permit. Clear guidance
•hould be given as to what is required,-how it should be documented, when it
,hould be delivered, and where the documents should be filed. Ideally, as-
)uilt assessments will follow immediately upon completion of a project. How-
>ver, given the variability in scheduling the phases of a wetland project (e.g.,
tesign, excavation, planting), as-built assessments may not be completed until
nonths or years after construction. To assist future evaluators, record the ap-
jroximate time elapsing between completion of the project and the com-
nencement of monitoring. Although the as-built assessment should be con-
iucted at the time the project is completed, delayed data collection,is
referable to no assessment at all.
The effort required to produce an as-built description of a site will depend
)n how closely the construction plans were followed. If the wetland was con-
ducted as planned, very little may be required other than to verify that the
)lans are correct (e.g., the original site maps will not need to be redrawn). If,
towever, construction differed from the plans, actual site conditions must be
\n Approach to Improving Decision Making in Wetland Restoration and Creation
52
-------
AS-BUILT CONDITION FIELD MAP
C2-T1
July 1987
Site
Photo 3
Land Use
Photo 1
VT2
VT1
Site
Photo J
BMT1
Site
Photo 1
VT Vegetation Transect'•' r -
BMT-*- Basin Morphometry Transect
A Start of Transect
H End of Transect
Culvert
— Pond (open water)
•'^'5"-2>Sr£' Planted Emergent Vegetation
..JSurvey. (anticipated wetland
boundary)
1:500
O 5 10
m
-IV-
0 21 42ft j|
Wetland Area = 0.3 ha
Pond and emergent vegetation boundaries are the approximate locations observed in July, 1987.
Data collected by Stephanie Gwfrrand Sheri Confer.
Map drafted by Tracy Smith.
Figure 4-1. Example of a Field Map to document as-built conditions of a wetland project.
Chapter 4: Monitoring Performanc
53
-------
GO
5
CO
'
I
f-
o
5
c
s
CO
I
£3
,
£
-a =
J
o .
_u "R
Is
a. ca
Cf
•*
f
In Approach to Improving Decision Making in Wetland Restoration and Creation
54
-------
Watershed
Boundary
Figure 4-3. Map enlarged from U.S. Geological Survey quadrangle showing drainage area,
surrounding land-use, and wetland location.
Chapter 4: Monitoring Performance
55
-------
fully documented. This will require additional mapping and data collection
^nce^ls^uilt assessment is complete, differences between what was
planned and what was built should be evaluated by the permitting agency or
Organization responsible for the project. If modifications to the project are nee-
e&ry at this point, the as-built assessment will need to be updated to reflect
lese changes When it is final, file the as-built assessment with the perma-
nent project records so that it is available for comparison with future site-as-
ru, ncuu.a, wetlands, the initial assessment will consist of a base map that
documents the conditions found at the site and a supporting narrative Subse-
q\ient assessments will rely upon an accurate portrayal of these initial condi-
tions. :
' ' • •. '.' '. i . . •• :', " 1 -",1
ROUTINE ASSESSMENTS „ .
, . Routine assessments are simple site examinations or "spot checks used to
monitor and record wetland development. During the check, visual assess-
ment of wetland conditions are compared to maps and photographs from
prior visits and the differences noted. This information is used to: 1) identify
problems that require correction; 2) provide a record of progress; and 3) deter-
mine, in some cases, when site performance warrants releasing the contractor
from further responsibility.
)*oi '
Routine assessments, are less costly and potentially less damaging to the
site than more comprehensive assessments because data collection is less in-
tensive and many observations can be made without entering the interior of
the wetland (Kadlec 1988), Depending on the size and complexity of the pro-
ject or the natural wetland being monitored, routine assessments generally
take less than a day to complete. They require little equipment, and limited
training of personnel. '
Conduct routine assessments during the first few years after, wetland con-
struction when plant communities and hydric soils are becoming established.
Comprehensive assessments may be unnecessary or inappropriate until the im-
mediate-effects of construction activities have passed. Data from prior studies
of natural wetlands or wetland projects can be used to determine if and when
routine assessments should be replaced or alternated with more comprehen-
sive assessments. The time required for wetland projects to develop such spe-
cific attributes as stable or mature vegetation communities or wildlife habitat
can be estimated from these earlier studies.
An Approach to Improving Decision Making in Wetland Restoration and Creation
' , ^' 56 . ' ' . ..
, , ; j • ; i
-------
What to Include
Data collected during routine assessments should reflect project objec-
tives. At minimum, routine assessments include the collection of the types of
data noted in Tables 4-1 and 4-2. This set of standard data guilds upon data
collected during the as-built assessment, as shown in Figure 4-4, and can con-
tribute to a regional database on wetland performance and design (see Chap-
ters 5 and 6). . . . . .
The decision to collect certain types of data is made on a project-by-pro-
ject basis, and may result in the use of methods from both routine and compre-
hensive approaches. For instance, if sediment retention is a stated objective,
the routine assessment would include at least a visual inspection of water :flpw
rates and patterns and any associated evidence of sedimentation. Evidence of
an alluvial fan at a wetland's inlet may indicate excessive sedimentation. If
more accurate sedimentation data are required, then more quantitative meth-
ods such as annual measurements of sediment accumulation on feldspar clay
pads may be warranted (Cahoon and Turner 1989; Barbara Kleiss, COE, Wa-
terways Experiment Station, Vicksburg, MS, personal communication).
Routine assessments should be repeated at appropriate intervals to deter-
mine if the project is on track and objectives are being met, and should be per-
formed during an appropriate time of the year. For instance, if a project objec-
tive is flood peak reduction, inspect the site during flood events to see if it
receives floodwaters. A comprehensive assessment of flood storage function
might involve calculation of the actual volume of water stored (e.g., Simon et
al. 1988). Similarly, objectives relating to wildlife use suggest inspecting the
site during breeding, nesting, or migration seasons. Inappropriate timing of
wetland visits can lead to high variability in the data. Alternatively, high vari-
ability in data collected from different routine assessments may indicate that
another indicator should be used to assess the wetland function being studied.
In addition, some variability will be due to natural changes in the wetland and
are to be expected.
Generally, perform routine assessments annually or until you are confident
the project is developing as expected. This allows major problems (such as
excessive sedimentation or failure of a water control structure) to be identified
and corrected expeditiously. Regular annual checks also provide information
on the wetland's structural development and functional performance over
time. This is essential for determining if performance criteria are being met (see
Chapter 5). The information obtained is also needed to establish or evaluate
performance criteria for all wetland projects in the region^ As this information
accumulates in the project record, the frequency and timing of assessments
can be modified, as necessary, to produce reliable data. ;
In summary, annual routine assessments continue until the objectives of
the project are met and the contractor Is released from contractual obligations,
Chapter 4: Monitoring Performance
57
-------
ROUTINE ASSESSMENT FIELD MAP
C2-T1
June 1990
Land Use
Photo 1
Site
Photo 3
Site
Photo 2
"LjS**8^f*. .^ ,>*. >^ 7:^ ^ jKt ,j- ^., ^i... • ^»*
•*>? x •*% ^-./T>ff^ *~ j«5L >• >• »•
VT1
Surface Water
(Increased by
20%)
VT
Vegetation Transect
-*- Basin Morphometry Transect
Start of Transect
• End of Transect
Culvert
Pond (open water)
rS' Planted Emergent Vegetation
Survey (anticipated wetland
boundary) .....
Typha latlfolia
1:500
0 5 10
m
-N-
h
0 21 42ft
Wetland Area = 0.25ha (-17%)
Pond and emergent vegetation boundaries are the approximate locations observed in July, 1987.
Data collected by Stephanie Gwin and Sheri Confer.
Map drafted by Tracy Smith.
Figure 4-4. Example of a Field Map to document conditions found during a routine assessment
as compared to the as-built condition. Heavy dark line indicates most recent wet-
land perimeter and separates areas of dominant vegetation types. Note change in
wetland shape as compared to as-built conditions shown on Figure 4-1.
An Approach to Improving Decision Making in Wetland Restoration and Creation
i i ji i
58
-------
OR until the routine assessments are replaced by more comprehensive moni-
toring procedures OR until experiences indicate that projects at a certain stage
of development need less frequent assessment. Even after contractual obliga-
tions are fulfilled, continued periodic, routine checks can provide important
information on the wetland's persistence and performance over time.
A report should be compiled after the routine assessment is performed.
The report clearly indicates if corrections are required or if more comprehen-
sive monitoring is needed to interpret wetland conditions. : The report also
documents significant changes at the site that have occurred since the as-built
conditions were documented or the last routine assessment was performed.
The completed routine assessment should be furnished to managers at the
permitting or sponsoring organization for evaluation. The report can be used
by managers to make decisions such as requiring more comprehensive assess-
ments, continuing routine assessments, or releasing the contractor from further
responsibility. The routine assessment should be filed with the permanent pro-
ject records so that it is available for future reference, and appropriate summa-
ry information entered into any associated database to keep reports on the pro-
ject current.
COMPREHENSIVE ASSESSMENTS .- - ;
Comprehensive assessments generate more complete and quantitative in-
formation on the wetland's"performance, than, do routine assessments. Infor-
mation gathered during comprehensive assessments is important to: 1) identify
modifications to the site that are required to .meerproject objectives; 2) pro-
vide a basis for evaluating'projectdesign and establishing performance crite-
ria; 3) help explain why a wetland project was or was not successful, and 4)
support long-term research efforts.
Rationale
Comprehensive assessments are generally more costly, require more
skilled personnel, and can result in greater disturbance to theisite, as it is nec-
essary to sample and operate in the interior of the wetland.. Therefore, they
should not be performed until substrates"have""sla|inzed and plant communi-
ties are flourishing. The only exception would be to meet the needs of re-
search efforts to evaluate the early development of sites,. Comprehensive as-
sessments vary in breadth, detail, and frequency of data collection depending
on project objectives and thei needs of.the sponsoring organization.
--Comprehensive assessments should generally be performed when suffi-
cient time has elapsed after wetland construction to allow major wetland char-
acteristics to develop. This may be three to five years for emergent wetlands
and longerfo'r forested wetlands. S6metimes> however, specific project condi-
tions require a thorough or partial in-depth assessment at an earlier stage. An
Chapter 4: Monitoring Performance
59
-------
example of this is the protection or enhancement of an endangered plant
species which requires careful early monitoring of the species' condi-
tion throughout several growing seasons. Comprehensive assessments may
also be needed if a routine assessment indicates possible problems with the
site and additional information is required to determine appropriate corrective
actions.
What to Include
There are no specific procedural requirements for comprehensive assess-
ments because the reasons for conducting them vary. Tables 4-1 and 4-2,
however, list some possible procedures and why they would be used, and pro-
vide additional sources of information on methods.
Data collected during each level of assessment must be compatible with
data collected previously. Although data collection and analysis should be of
the highest possible'quality during all levels of assessment, it is particularly im-
portant for the comprehensive assessment because of the effort expended at
this level. The rationale used to justify an intensive sampling effort includes
specific objectives framed as hypotheses so that defensible conclusions can be
reached, In addition, to meet quality assurance objectives, the reasons for col-
lecting data on the chosen set of variables must be carefully thought out and
documented. We recommend: 1) development and evaluation of standard
operating procedures and sampling protocols by knowledgeable individuals;
2) acknowledgement of possible sources of error and bias in the procedures;
and 3) collection and evaluation of quality assurance replicates during all
phases of field and laboratory work to maintain scientific defensibility.
Copies of procedures, data, and assessment results should be supplied to
the organization responsible for the project. This material is then filed with the
permanent project records and becomes available for future site assessments
or research to help maintain consistency overtime and improve the interpreta-
tion of results.
Part of a comprehensive assessment is an analysis and evaluation of the
wetland's development and functional performance over time, based on com-
parisons to as-built conditions and previous routine arid comprehensive assess-
ments. The current status of the wetland is determined with respect to intend-
ed type and area (as required by the permit conditions or project objectives),
and its sustainability as a functional system in the landscape. If design correc-
tions are needed to meet project objectives or to maintain the system, the pos-
sible impacts of modifying the existing conditions on the site must also be con-
sidered. Alternatively, you may need to reevaluate and adjust the performance
criteria required to more realistic levels, given the uncertainty of wetland
restoration and creation technology and variations in environmental condi-
tions from year to year.
An Approach to Improving Decision Making in Wetland Restoration and Creation
60
-------
ASSESSMENT VARIABLES
The variables measured during monitoring at the various levels are dis-
cussed in the following sections. We are suggesting methods that we have
used successfully in the field. . •
We realize, however, that other methods may work equally well. See
Horner and Raedeke (1989), Adamus and Brandt (1990), and PERL (1990) for
additional recommendations.
General Information
Standard information must be collected to identify the location of each
project and natural wetland being monitored. The project or permit file will
contain much of the required information, but often it will need to be amend-
ed during the assessment of as-built conditions to include .a narrative descrip-
tion, photographic record, and, most importantly, an accurate map. Because
the condition of a wetland often depends upon its surroundings, we recom-
mend determining its position in the watershed (e.g., headwater, stream order,
floodplain, isolated), and measuring the receiving drainage area on a topo-
graphic map (Figure 4-3). Obtain the watershed boundaries or drainage area
from a USGS (1:24,000 scale) quadrangle, unless the wetland is quite small, in
which case a survey done in the field may be substituted. Classify both the
wetland type and the surrounding land use-according to standard systems,
, such as Cbwardin et al. (1979) and Anderson jet al. (1976). Then use the map
to estimate and record the percentage of each land use type occurring within
at least a 300-m band around the ,wetland,(Table 4-2, Figure 4_-3), Routine as-
sessments can record observations and changes on theiase map created dur-
ing the as-built assessment (Figure 4-4), to provide consistency over time and
reduce the mapping effort during subsequent visits. •
Morphometry
Most wetland projects, and many natural wetlands are located in topo-,
graphic depressions or basins. Measurements of physical features; such as
area, slopes, and water depths should be made during all assessments and
used to construct a. map from an aerial-view and topographic profiles from a
side view (Figures 4-1 and 4-2). Accurate portrayal of the wetland in its as-
built condition is particularly critical, since this will form the basis for future
comparisons. These data can be collected along transects as can data on sub-
strate, vegetation, hydrology, and fauna (Figure 4-5).
.".A site location map should identify watershed position and land use adja-
cent to the site (Figure 4-3). If accurate as-built maps are available, only limit-
ed field work will be needed to complete the assessment. If information on
the as-built condition is not available, thenjQeld mapping must occur during
the first monitoring visit to the site. Use standard wetland delineation proce-
Chapter 4: Monitoring Performance
61
-------
II ..... in1!: ........ !, i| •' i ,
HI! .......... l"!i ...... I !'"«" 'is ....... FIT1;1 nil1: ....... . " ...... in: • IP 'if f i
Figure 4-5. Field crew members taking elevation measurements along a transect.
An Approach to Improving Decision Making in Wetland Restoration and Creation
62
-------
dures to determine the extent of wetland present. Keep in mind that newly
constructed wetlands may not have developed all the characteristics necessary
to meet the criteria for a jurisdictional wetland. Therefore, it may be necessary
to estimate the wetland area based on slopes, planting patterns, and existing
hydrology. The map can be modified during future assessments. Although
physical morphometry of the wetland is unlikely to change dramatically over
time, the jurisdictional boundaries may fluctuate.
Hydrology i
Fluctuations of water level and the duration of inundation or saturation de-
termine, in part, the composition of plant communities (Erwin 1988). Inunda-
tion of water at the surface can be easily observed and recorded on a map.
There are times throughout the year, however, when site visits will not coin-
cide with surface inundation, and when soils are saturated below the surface.
Therefore, several shallow wells are commonly installed within a wetland to
measure water levels below the surface to depths of 0.5 to 2.0 m, depending
on the expected movement of the local water table. Plastic (PVC) pipes 50-75
mm in diameter with narrow, horizontal slots were used.successfully in nu-
merous projects. These pipes can also be used to measure depths of surface
water when standing.water is present, or separate staff gauges can be installed
(Horner and Raedeke 1989).
During site visits, describei and" re cord on a map the flow rates and patterns
of surface water. For wetlands.with distinct inlets or outlets, flumes or weirs
can be used to measure discharge. Locate any water control or containment
structures on the map» and describe :them as well._ Document and photograph
hydrologic indicators as described Tn the Federal ICWD (1989), e.g., drift lines,
water-stained leaves, oxidized root channels. Single monitoring visits during
the year are not likely to yield reliable information about wetlands with vari-
able hydroperiods, so we recommend multiple visits to make readings during
several seasons.
Substrate
Substrate characteristics often reveal hydric conditions. Characteristics
such as soij color and moBing^ whicti^ndjcate the duration and depth of soil
saturation, can be determined quickly and require little training for evaluation.
Gleyed soils (those predominantly neutral gray in color and occasionally^
greenish or bluish gray) are typically hydric. Mottle abundance,, size, and
color usually reflect the duration of the saturation period and indicate whether
the soil is hydric (Federal ICWD 1989). Use a Munseil colpr chart to deter-
mine the hue, value, and chroma of both the mottles and the surrounding soil
matrix during as-built and comprehensive assessments (Figure 4-6). The per-
cent soil organic mltter determines the suitability as a planting and growth
.-.-•-.- • --•• .. Chapter 4: Monitoring Performance
63 '
-------
, i (. Ill
l-lt
Figure 4-6. Field crew members using a Munsell color chart to determine soil hue, value, and
! chroma., . '!...,'. • . '. :". , .-.,:•'", .."i"".
medium. The proper percentage of organic matter and the proper soil texture
and "hardness" are required to allow penetration by roots and rhizomes (Owen
et al. 19&9) for vegetation establishment. Soil organic matter also provides
necessary nutrients for microbial activity. Refer to Chapter 6 for a more de-
tailed discussion of soil organic matter and substrate characteristics.
.!!, " ' ',-.,., :
Vegetation '
'A species list and the arrangement of plant communities on the site are
commonly used to characterize the vegetation of wetlands. Vegetation data
can be collected in a variety of ways (e.g., Brower and Zar 1984-, Pielou 1986).
For documenting as-built conditions, the identification, coverage, and location
of each planted species is essential. These data serve as a bench mark from
which' to compare the plant community as it matures or changes over time and
to determine the survival of plantings. For routine assessments of projects or
initial visits to. natural wetlands, visual estimation of the percent"plant cover
and a list of dominant species is usually sufficient. However, comprehensive
assessments require a more quantitative approach using quadrats of various
Sizes (circular, square, or rectangular in shape, 0.1, 0.25, or 1.0 m2 for herba-
ceous plants, >1.0 m2 for shrubs and trees, (Brower and Zar 1984, Horner and
Raedeke 1989, Leibowitz et. al. 1991) to measure density or coverage (Figure
4-7). The size and shape of quadrats chosen for sampling should be appropri-
Ah Approach to Improving Decision Making in Wetland Restoration and Creation
• • ' " 64
-------
Figure 4-7. Botanist reading a vegetation quadrat.
, Chapter 4: Monitoring Performance
65
-------
1 • ,,l !
ite for the type of vegetation being sampled, remain constant over time, and
)e consistent among sites.
Once the extent and boundaries of the plant communities are mapped,
ind a species list is generated, these data can be used to evaluate wetland de-
jendency (Reed 1988, the appropriate volume; Wentworth et a!. 1988) and
>tfier characteristics of the plant community (e.g., ratios of native to exotic
;picies; see Chapters 5 and 6). The permit conditions for wetland projects
ypically focus on the community composition, coverage, and survivorship of
jlants, so monitoring efforts, are, in part, directed at obtaining these data.
Annual routine assessments generally are performed during the mid-to-late
jrowing season for most wetland types (Brooks 1990), although sampling dur-
ng other seasons may be more appropriate for certain types (e.g., vernal
aqpls). One factor to consider when choosing a time to sample is that the
ivailability of mature fruits helps in the identification of plant species. In addi-
jon, wetland plant communities may be absent or hard to identify at certain
jmes of the year. Check ephemeral wetlands such as vernal pools and some
/vet meadows when wetland vegetation is present and fruiting, even though
lydrologic evidence may be lacking at that time. Although the plant commu-
nity is usually sampled during the mid-to-late growing season when species
:omposition is of primary concern for the wetland in question, multiple visits
•nay be warranted due to changes throughout the growing season.
• Fauna ' ' -''.'•'•: -. ..-'••-. -."..-'".'Zr."' " -•."-.
the habitat value provided to wildlife and fish is frequently cited as a
najor wetland function and objective of projects. The use of wetlands by di-
/erse faunal communities has influenced both wetland protection arid'man-
agement. Relatively few studies have monitored the diverse fauna that use;
/vetlands (e.g., Brooks and Hughes 1988, Brooks et at. 1991), so sampling pro-
:ocols and comparative data are relatively scarce as compared to the literature
ind data on plants. If one objective for a. project Is to create habitat, then
some assessment of habitat condition must be included in the sampling proce-
dures. Minimally, direct and indirect (e.g., tracks, "sc'at) observations of verte-
Drates and invertebrates must be recorded during" all'levels of assessment.
vtpre quantitative Information can be gained by evaluating habitats for a few
selected indicator species (e.g., wood frog for northeastern forested wetlands)
jsiirig the Habitat Evaluation Procedures (HEP) developed by the USFWS
1980). HEP is most applicable for temporal comparisons as the vegetation on
he study site matures, or for comparisons between projects and natural wet-
ands. As required, specific census techniques can be used to determine the
Dresence and abundance oFselected faunal groups (Schemnitz 1980, Erwin
1988, Homer and Raedeke 1989, Brooks et al. 1991, Murkin 1984) (Figure 4-
3). The timing of censuses can significantly affect results, so we recommend
\n Approach to Improving Decision Making in Wetland Restoration and Creation
66
-------
Figure 4-8. Field crew member collecting invertebrates from an emergence trap.
monitoring the fauna! community with careful attention to expected daily and
seasonal variations (e.g., early morning surveys for birds and spring breeding
surveys;for_imphibjans).,.."...„_.-:.—_^i:_-.v.:'x~iv-" -";^-v ;. :.-;
Water Quality
We suggest avoiding a substantial investment in analyzing water samples
unless the project of interest has water quality improvement as a primary ob-
jective (e.g., constructed to retain nutrients or treat storm water) or a specific
pollutant load is expected (e.g., heavy metals or pesticides are found in high
concentrations in the adjacent landscape). Implementing a water quality mon-
itoring program for a large population oi wetlands may be prohibitively expen-
sive. An additional complication is the inherent spatial and temporal variabili-
ty found in the chemical characteristics of water in wetlands. The choice of
water quality parameters is left to the discretion of the investigator. See Horner
and Raedeke (1989) for suggestions on how to implement a water quality
monitoring program for wetlands. —..-•.-
Additional Information .
Finally, all levels of assessment should include a photographic record and
descriptive narrative of conditions present in the wetland and surrounding
-•--•••-- - Chapter 4: Monitoring Performance
67
-------
landscape (Table 4-2). In addition to providing a historical record of visual
changes, a photographic record allows you to begin your evaluation of a site
ih.the office. Photographs are no substitute for quantitative data, but are an in-
vkluable aid to documentation of site conditions.
When taking photographs, specify a standard protocol to facilitate inter-
site and temporal comparisons. We suggest using a 35-mm camera and 50-
rrim lens. Color film is recommended; prints (and the accompanying nega-
tives) provide a more convenient format for a permit file, but slides are also
useful. Take sufficient numbers of photographs to allow evaluation of the site
from all directions. Consider taking photographs from a permanent station
along the four major points of a compass. Indicate the locations of photo'sta-
tions and the directions of photos on the site map.
A brief narrative should be included to describe features or findings that
do not fit into the above categories (e.g., the water control structure was van-
dalized resulting in a reduction of water depth), and to document the current
condition or observed changes from past assessments (e.g., the water control
structure should be repaired because the drop in water level has decreased the
wetland area by 20%).
DEVELOPING AN EFFICIENT SAMPLING STRATEGY
, The cost of post-construction assessments varies dramatically with the
methods and intensity of data collection. Given the potentially large number
of wetlands in a target population, this cost must be balanced against the value
of the information collected. Devoting a high level of effort and expense to
data collection is neither appropriate nor necessary for all assessments. To re-
duce costs and increase accuracy, select cost effective and efficient assessment
methods (e.g., PERL 1990).
One way to increase assessment efficiency is to use compatible field meth-
ods, units, data analysis, and reporting procedures. This promotes more accu-
rate, cost effective, and meaningful post-construction assessments. A standard-
ized data form designed for ease of data entry facilitates development of
consistent assessment procedures and allows the aggregation of data from
many wetlands, which in turn, enables the development of a regional wetland
database.
. Establishing permanent sampling plots is another way to make different as-
sessment levels compatible. Visits to permanent plots are not likely to signifi-
cantly affect the wetland for any of the three levels of assessment proposed. If
permanent plots are used, however, sampling must be nondestructive and
using various routes to access the plots may be adviseable to avoid creating
paths. For example, plant specimens should not be removed from plots unless
absolutely necessary to avoid introducing a possible bias in species occur-
rence or abundance. '
An Approach to Improving Decision Making in Wetland Restoration and Creation
68
-------
Agency personnel typically must evaluate multiple sites wjthin a short time
period. A schedule designed to allow sampling of sites in geographic proximi-
ty'will reduce the time spent traveling to sites, thus improving sampling effi-
ciency and reducing costs.
Another method for reducing cost is to use local volunteers. If properly
trained and supervised, volunteers can be a source of high, quality assistance.
The special insert following this chapter discusses the role of local volunteers
in site assessments. . , • .
The sampling strategy determines when, where, and how to collect data.
Your strategy will vary with the project goals and the specific assessment level
.being performed, but the sampling methods should remain consistent. This
promotes data comparability among sequential assessments. For instance,
vegetation assessments should be performed in plant communities during the
same part of the growing season each year rather than at the same date be-
cause of differences in weather year to year. The ability to compare results of
different assessments over time enhances your ability to evaluate performance
criteria as wetland restoration and creation technology improves.
Data Quality
It is important that the field methods selected provide high quality data
that are scientifically defensible. Variation in data due to sampling, collection,
and processing methods must be as low as possible, or actual changes in site
conditions may not be detected. In our studies, we have considered five basic
quality assurance components: precision, accuracy, completeness, representa-
tiveness, and comparability. Each component addresses a different aspect of
data quality.
Precision is a measure of mutual agreement among individual measure-
ments of the same variable, usually under prescribed similar Conditions (Sher-
man et al. 1991, adapted from Verner, 1990). Precision is usually expressed in
terms of the standard deviation, however, precision is calculated differently
depending on the variable and method used for measurement. You can use
field and laboratory duplicates, standard procedures, ancl process repetition by
separate individualsto-achieve better precision in your data.
Accuracy is the-degree to which a measurement represents the true or ac-
cepted reference value of the variable measured (Sherman et al. 1991). Accu-
racy depends oh the technique used to measure-the variable and the care with
which it is executed. It is difficult to assess accuracy for many field measure-
mentsrhowever^ examples of ways to improve the accuracy of your data in-
clude: use of tested standard procedures/training of field crews, and use of
standard reference materials. i"
Completeness is a^measure,bf the amount of valid data obtained compared
to the amount that was expected under ideal circumstances (Sherman et al.
Chapter 4: Monitoring Performance
69
-------
1991). You may not be able to collect all the expected data due to time con-
straints, adverse field conditions, or sample and data loss. If the number of
samples taken is less than that originally intended, seek the advice of a statisti-
cian to make sure that the sample size is large enough to produce data that ad-
equately represent site conditions. Proper sample and data handling proce-
dures (e.g., labeled samples and legible data forms) can reduce the chance of
lost information.
Representativeness expresses the degree to which data accurately and pre-
cisely represent a characteristic of the variable of interest (Sherman et al.
1991). Consider representativeness during site selection to ensure that the
sites chosen are representative of the population (See Chapter 3). Transect and
plot establishment should also represent typical conditions of the site being
sampled.
Comparability expresses the confidence with which one data set can be
compared to another (Sherman et al. 1991). By comparing duplicate data col-
lected by different field personnel working on the same sites and plots, an esti-
mate of data variability caused by individual bias can be obtained. If the vari-
ability of data collected by different individuals exceeds the inherent
variability of the measurements of the same variable, then the sampling strate-
gy will require modification. Field crew training and standardized procedures
will improve data comparability.
Where to Collect Samples
Sampling strategy depends on the variables being measured, their distribu-
tion across the site, and the intended use of the data. For instance, water sam-
ples collected to provide information on how a wetland functions as a nutrient
sink need only be collected at water inlets and outlets.
Wetlands frequently have heterogeneous distributions of vegetation and
soils. These distributions can be sampled with a systematic (i.e., samples are
taken at predetermined intervals) or stratified random (i.e., samples are taken
randomly within subdivisions of jhe unitlaeing sampled) sampling design.
Transects for systematic sampling should be established parallel to environ-.
mental gradients such as moisture or elevation. If natural variation is not asso-
ciated with an identifiable environmental gradient, collect samples randomly
or systematically from each major stratum so that the full range of the
variable's attributes is represented in the data.
How Many Samples to Collect
Both the cost of collecting data and the potential site damage due to tram-
pling during field work increase with sampling intensity. Sampling intensity,
however, must be high enough to produce data that adequately represent site
conditions. High variability within the site increases the number of samples
An Approach to Improving Decision Making in Wetland Restoration and Creation
70
-------
required. The tradeoff is cost and site damage versus data precision. Kreb:
(1989) is a good reference for various methods of determining the optimurr
number of samples to collect for ecological studies, as are most biometrics o
statistical reference books (e.g., Snedecor and Cochran 1980).
When to Collect Samples
The assessments discussed in this chapter serve different purposes an<
thus, the timing and frequency of sample collection will vary. In general, sam
pling should be timed to match important phenomena relating to the projec
.objectives (Brooks and Hughes 1988, White et al. 1990,:PERL 1990, Leibowit;
et al. 1991). As noted, wildlife habitat use should be checked during the timi
the target species is likely to occupy the wetland, and vegetation should b
checked during the part of the growing season when plants can be most easil;
identified. Hydrologic sampling in a region with evenly distributed patterns c
precipitation throughout the year will differ from that of an arid region or oni
with distinct wet and dry seasons. ,
Sampling frequency required depends on the preconstruction conditions a
the project site (White et al. 1990). The probability of successfully restoring;
wetland often depends on the extent to which the wetland is degraded. I
most of the attributes of a functioning wetland (e.g., hydric soil, wetland vege
tation, and correct hydrology) are still present or easily repaired, the project 5
more likely to succeed. On the other hand, if major wetland attributes hav
been destroyed, or. if a wetland is being created on an upland site, the succes
of the project becomes more uncertain. Projects with very uncertain outcome
_ require rnqre frequent and/or intensive monitoring sojthat timely design coi
rections can be made if necessary (White et al. 1990).
Controlling Damage to the Site
Performing site assessments can damage a developing wetland. Therefore
post-construction assessments should be designed to minimize activities with!
the wetland. It is best to minimize walking within the site:during reconnah
sance and sampling activities. Approach permanent sampling transects c
pointe-by: alternatejoutes during successive visits. Do not traverse the sam
place repeated[ylto.ayoid:developing a trail. Trails, besides destroying vegetc
tion and altering water flow, are often invitations for other people to enter th
wetland, sometimes on horseback, motorcycles or off-road vehicles.
SUMMARY" ":"' ----••-"•----- '--^--- ---. •••;-- --'--:
Insight into the probable success of wetland restoration and creation e
forts will enable you to make the sound wetland management and policy dec
sions required to-protect the resource. Post-construction wetland assessmen-
Chapter 4: Monitoring Performan<
71
-------
re essential to confirm compliance with permit conditions or project objec-
ves and to ensure that projects provide the functions expected.
The assessment procedures used in our Approach are selected to enable
corporation of the data into a regional database that characterizes both wet-
md projects and natural wetlands. The database will provide a basis for refin-
ig regional wetland performance criteria and design guidelines and identify-
ig design characteristics that are most likely to produce the desired functional
>sults. In addition, recording performance levels of wetland projects pro-
iptes the establishment.of. appropriate and attainable wetland performance
riteria.
ICI Id. , ..-'.'
Assessment procedures should produce high quality, cost effective data.
.dequate information must be collected to assess and promote project suc-
ess, and to help refine regional mitigation policy. At the same time, assess-
lent activity costs must be minimized to reduce the financial burden on both
ie public and private sectors. . .
Using the WRP Approach, a monitoring plan based on project objectives
; established as an integral part of the wetland project. Assessment proce-
ures, an<^ timing and frequency of sampling are clearly documented. .The
Ian also identifies the person or organization responsible for performing the
ssessments^ who is to receive and archive the reports, and who is responsible
3r performing and overseeing corrections. Because corrections can be costly,
strategy, for evaluating the assessment findings is included as part of the plan.
To effectively protect wetland resources, we must identify which wetland
/pes and functions can reliably be replaced in a given time and geographic
sgion. We heed information on how to manage wetlands, on which wetland
esigns work, and on how restored and created wetlands are likely to persist
nd function'over time. A well planned monitoring program plays an essential
in acquiring this knowledge.
n Approach to Improving Decision Making in Wetland Restoration and Creation
„: • "i' ' ' ' ' i. 1 -
72
i'l •! '< ''Ill •'.. ' •' I . 'I'1' "" I.'.' " . ". . ' .'ij :-• . V .'l|l|li !:.| . , ,. . | i.;, ,
-------
VOLUNTEERS AND NATURAL RESOURCE
MONITORING
by Neal Maine
Editor's Note: We are pleased to offer the following special section as an ex
ample of how a monitoring program can be implemented by an agency de
spite limited staff and funding. Author, Neal Maine, exemplifies the positivt
impact one person can have on protecting coastal resources. He won the pres
tigious Chevron Conservation Award in 1988 after being chosen to represeh
Oregon by then Coverner Neil Goldschmidt because "he educates Oregon')an
about our coastal resources by encouraging citizen involvement in conserva
tion projects". Neal's contribution spans 26 years as a teacher in the Seasidi
Schools and as an instigator and volunteer in north coast wetland conservatioi
projects. Asked how he has managed to achieve so much, Neal says: "Whei
I see a need or where something is being threatened and nobody is doing any
thing about it, I just try".
VOLUNTEERS AND ENVIRONMENTAL
PROTECTION
Protection of the environment directly af
fects quality of life values. For this reason, en
viron mental issues are often the focus of com
munity action. Whether it is a group effort t(
preserve habitat for a valued species or t<
clean up a polluted area, many people are be
coming more involved in protecting an<
restoring their environment.; Federal and stat<
agencies and nonprofit organizations promoti
_ _= _ involvement by volunteers in environmenta
activities because they know how essential local support is to the ongoing sue
cess of any project; Local participation includes activities such as communit;
education, helping with work projects, building trails, recording observations
and ."adopting" resources as varied as trees, streams and animals.
In the state of Washington, the Adopt-A-Stream Foundation invite
volunteers to adopt a stream or wetland, collect data about thi
Volunteers and Natural Resource Monitorin
73
-------
resource, and report the data to a planning organization for use in
making decisions about needs for resource management.
• Volunteers work with the Oregon Department of Fish and Wildlife to
conduct salmon spawn surveys, rehabilitate streams, and incubate
salmon in streamside hatchboxes to help sustain Northwest salmon as
both food and game fish.
• The National Wildlife Federation encourages individuals and groups
to assist with winter bald eagle counts to ensure that the eagle
population is not diminishing.
The Nature Conservancy invites individuals and groups to participate
, in stewardship programs at selected reserves.
• Cannon Beach, Oregon, community volunteers educate visitors to the
Haystack Rock tidal pools so they can enjoy the area without harming
the pools or their inhabitants.
.{"! ''!; r , • • . , ' -:• +• , I , !
"Til" "J1!! '.. ! I" -. ,•": ': ; ",";•••• •,» ; i] ;:J(T : If • ' ,1" [;•!, •',• : ; |
;•: The Environmental Protection Agency (EPA) has just completed a
program called Streamwalk in Region 10 in which volunteers work
: with scientists to collect genera! information about a stream by
mapping surveyed areas, characterizing the stream, and determining
the cause of any adverse conditions found in the stream.
•!• .!• • . -. r • - 't I : , . r I
All of these programs involve citizens in the process of monitoring and
rotecting natural resources. When local citizens become involved with nat-
ral resources and their management, a new kind of protection emerges. The
wareness of the entire community is focused on the problem at hand, the en-
ironment becomes our environment and is afforded the respect ancl protec-
oh reserved for the Earth as our shared home.
MONITORING MITIGATION PROJECTS
Funding for wetland mitigation, under Section 404 of the Clean Water Act,
; based on the need to meet the legal requirements of proposals by develop-
rs, agencies, municipalities, and in some cases, individuals, which involve
/etland alteration. Support for on-site monitoring of created or restored wet-
inds is often minimal. It is usually limited to single annual visits to the site for
dbjective evaluation. The collection of quantitative data is seldom required,
(though resource managers know that if they are to answer broad questions
bout the success of mitigation in restoring wetlands or specific questions
bout individual project success, the collection of valid data must be an inte-
n Approach to Improving Decision Making in Wetland Restoration and Creation
:.. r '. ' , • . ' , 74 '
-------
gral part of all projects. A major limitation for including detailed on-site envi-
ronmental monitoring in mitigation plans is the cost of professional services for
collecting field data. Logistic problems such as transportation and scheduling
constraints can also limit the collection of data on daily, weekly or monthly
cycles This is particularly true when sites to be monitored are some distance
from qualified scientific staff. Providing opportunities for participation by vol-
unteers can account for significant cost savings in professional services and
travel expenses. More important, while helping collect specific data for a re-
search project, volunteers begin to understand research goals and gain insight
into the biological processes that occur in both created and natural wetlands.
A local group of educated wetland supporters is created along with the wet-
land.
Aldo Leopold addressed this type of research 50 years ago in an essay in
the Sand County Almanac on the development of the land ethic. He stated,
"The last decade, for example, has disclosed a totally new form of sport, which
does not destroy wildlife, which uses gadgets without being used by them,
which outflanks the problem of posted land, and which greatly increases the
human carrying capacity of a unit area. This sport knows no bag limit, no
closed season. It needs teachers, but not wardens. It calls for a new wood-
craft of the highest cultural value. The sport I refer to is wildlife (resource) re-
search." Leopold concluded that, "The more difficult and laborious research
problems must doubtless remain in professional hands, but there are plenty of
problems suitable for all grades of amateurs."
' The following sections discuss the concept of supporting an increased
level of community involvement in scientific research activities, one in which
local volunteers become an integral part of the effort. The discussion centers
around how to build a research team that includes volunteers, using as an ex-
ample the Trail's End wetland creation project in which six amateurs became
part of a wetlands research team in Seaside, Oregon. Wetland research is the
focus of this particular volunteer project, but local volunteers can be success-
ful coworkers in most monitoring efforts.
THE TRAIL'S END PROJECT
" Comparing monitoring datagram natural wetlands and wetland projects
enables scientists to set standards for restored or created wetlands based on the
quality that is attainable in that region and ecological setting. For example, sci-
entists can identify when developing plant communities do not match the nat-
ural assemblages, and recommend corrections to project design based on typi-
cal local wetlands. The Trail's End study, conducted in the spring of 1989 in
northwest Oregon, was designed to use volunteers to collect extensive data
for a detailed monitoring study of a 15 acre created wetland, Trail's End, and
three natural marshes. One of the objectives of the project was to test the use
_T, . Volunteers and Natural Resource f^onitoring
75
-------
)f volunteers for collecting scientific data within the context of a wetlands re-
;earch project. Specifically, the objective directed researchers to evaluate the
jse of citizen volunteers in many aspects of the data collection.
Six volunteers from the Seaside community were recruited and trained to
vork with scientists en contract to the Environmental Protection Agency.
These local volunteers accepted the responsibility for collecting data on vege-
ation, water chemistry, emergent and benthic invertebrates, hydrology, soils,
irid bird use. They als3 participated in mapping the sites, developing a photo
•ecord, and testing metiods.
We learned from the Trail's End experience that the process of putting to-
jeiiher the local team should receive special attention. This team is the spon-
;oring organization's link with the data collection activities and the wetland re-
source. The final accjracy of the data, and therefore, the reliability of the
data, depend on the elfectiveness of the initial team-building and training ef-
:ort.
TEAM BUILDING
Individuals who choose to become volunteers are the key to the success of
i project, they are asked to put in many hours of work, often for long periods
Df time on a given day, and to collect data of a quality that will meet the stan-
dards of the sponsoring organization. In many cases, individuals are already
rivolved with issues, activities, or management related to wetlands and wet-
arid creation. These individuals are the best candidates for volunteers and for
:he role of Local Team Leader.
From the initial contact with the local community through the close of the
Droject, every effort should be made to create a cooperative working environ-
nent in which volunteers and scientists are part of a real team effort. To do
:his, sharing of information from goals and objectives to the final evaluation of
:he project and possib
/olved in planning the
sible changes; they oft
ocal resources.
We used a model
antists work cooperativ
sired attitudes and skill
e use of the results is essential. Volunteers can be in-
r part of the process, or in making decisions about pos-
m have creative suggestions and superior knowledge of
vith four interrelated roles in which volunteers and sci-
ely. Each role demands certain responsibilities and de-
PRINCIPAL INVESTIGATOR
Represents the
Responsible fo
sponsoring organization
the project's success
4/7 Approach to Improving L
ecision Making in Wetland Restoration and Creation
76
i
-------
• Proven expertise in the design, execution, and management of major
research projects
• Thorough knowledge of the specific wetlands ur ider study
Commitment to the concept of using volunteers
Good communication and interpersonal skills
OFF-SITE TEAM LEADER
Represents the Principal Investigator
Liaison between the scientists and the volunteer
Responsible for on-site procedural decisions
Thorough knowledge of wetland ecology
Experience in identifying wetland flora and fauna
Good communication and" interpersonal skills
LOCAL TEAM LEADER
• Represents the volunteers
Liaison between the Principal Investigator and
• (CRITICAL) Experience in organizing volunteers
Leadership skills for organizing and
cooperative team
• Commitment to the success of the project and eagerness to learn
* Good communication and interpersonal skills for working with local
agencies and the community
• Sound understanding of basic ecological concepts
• Knowledge of wetland flora arid fauna
in a research project
the
community
for research project
maintaining a productive and
Volunteers and
Natural-Resource Monitoring
77
-------
VOLUNTEERS
Should be from the local area
Willing to make a major commitment to the project
Interest in protecting wetlands, plants and wildlife
Some knowledge of natural history and science
Eagerness to learn how to perform assigned tasks
Ability to attend to detail and adhere to data collection schedules
• Ability to participate cooperatively in a team effort
The Principal Investigator and the Off-Site Team Leader are chosen by the
sponsoring organization. The Local Team Leader and the Volunteers must be
chosen with concern for both commitment and ability.
LOCAL TEAM LEADER
Following the designation of the Local Team Leader by the Principal Inves-
tigator ancl the Off-site Team Leader, a briefing meeting should be held to give
the on-site person a full sense of the project scope and the general nature of
the methods. It is critical at this time to establish the levels of responsibility for
the sponsoring agency, the Local Team Leader and the volunteers. The Local
Team Leader, who must be knowledgeable about community resources, will
make personal contacts to recruit volunteers. This individual's clear under-
standing of his/her role in providing leadership by coordinating the data col-
lection effort is essential for the success of the volunteer effort. It is also essen-
tial for the eventual validation of the field data collected and therefore, the
reliability of the project results. , An important factor in the success of the
Trail's End project was the Local Team Leader's skillful leadership, planning
and organization in determining the responsibilities of the volunteers and sup-
porting them in their efforts.
VOLUNTEERS
The Local Team Leader contacts individuals who might be interested in
volunteering for a community-based research project. It is helpful if they have
had some personal association with research, although not necessarily as a
professional scientist. These individuals are often found teaching science in
schools, studying science in college, and volunteering in natural history pro-
An Approach to Improving Decision Making in Wetland Restoration and Creation
78
-------
grams. Others have had previous experience in a science-related profession,
but are now working in different careers, and some have never had profession-
al or academic training in science, yet have developed their skills through per-
sonal interest. The most critical attributes of successful volunteers are commit-
ment to the success of the project; ability to attend to detail, adhere to data
collection schedules, and work cooperatively; and eagerness to learn.
The Local Team Leader develops a list of the prospective volunteers. After
the scope of study has been clearly defined, the Off-site Team Leader works
with the Local Team Leader to match the backgrounds and skills of the volun^
teers to the specific study tasks. Personal contact is generally the most effec-
tive way to make the final decisions. An open meeting in the1 community stim-
ulates interest and gives both team leaders a chance to meet with all of the
potential volunteers. The meeting should be structured so that there is time for
volunteers to introduce themselves, share their reasons for wanting to partici-
pate in the project, and work together in planning the next meeting around
their needs for formal instruction.
Both initial and closing interviews are an integral part of Devaluating use o(
local volunteers in data collection. Interviews with the research volunteers
who participated in the Trail's End project revealed that they were interested in
participating because of the chance to be a part of doing science, not studying
about it; a personal interest in wetlands and natural history; the opportunity to
use their skills and knowledge in a wetland setting; personal contact with pro-
fessional researchers; and the excitement of being part of a study on local nat-
ural resources. By the close of the project, these volunteers were also con-
cerned that without ongoing monitoring, local wetland mitigation projects
would not meet the public need for protecting wetland values. Three of the
volunteers at Trail's End were science teachers who transferred their skills tc
their high school students. The students are continuing to collect long-term
data using the field techniques from the original study.
TRAINING FOR VOLUNTEERS
After the team is identified, the next step ij
to complete jhe training that prepares the vol-
unteers and the. Local Team Leader to colled
data at a level that meets the standards for the
study established by the Principal Investigator.
The data not only have to be accurate, the>
also must meet the Quality Assurance (QA
standards of the sponsoring agency (EPA in the
case of Trail's End).
The introductory training session is the firsi
occasion on which all participants meet and i<
" Volunteers and Natural Resource Monitoring
79
-------
a good opportunity to begin building the team concept into the project. En-
couraging everyone to contribute ideas and ask questions, and making deci-
sions by consensus when appropriate, are two techniques for letting the volun-
teers know they are important to the project's success. In the first phase of the
Trail's End project, presentations by scientists on the research objectives of the
project, wetland ecology, data collection methods, correct use of reporting
forms, and a discussion of the components of QA helped volunteers under-
stand both the broad view of the project and the specific needs of the sponsor-
ing agency, training sessions were informal, with lots of time for volunteers to
ask questions about the project and to clarify any personal concerns about
participation.
Following initial training sessions, participants should be taken to the field
where they can get hands-on experience with each of the procedures to be
used in the project. For the Trail's End project, EPA staff presented instruction
about a task, then volunteers performed the task while scientists played the
role of QA observer. For example, using protocol instructions developed for
this study, volunteers conducted water quality tests for oxygen and pH. The
objectives were to compare water quality at the created wetland with the three
natural wetlands and to test the applicability of the chemical field kits. Results
were used to check the comparability of data between individuals and repli-
cates. Each volunteer had an opportunity to work through the procedure and
record data on the appropriate form. They received immediate feedback on
the accuracy of the data. The training staff was sensitive to the concerns of the
volunteers, who were allowed to repeat a technique until they were confident
in its use. All volunteers agreed that this was the most critical part of the train-
ing. The hands-on experience enabled them to master the techniques and un-
derstand the expectations of the EPA staff. Volunteers also had the opportunity
to change roles, from data collector to QA observer.
In some cases, such as the water quality sampling, data collection de-
pended entirely on the skills and knowledge of the volunteers. In others, the
volunteers supported the scientific staff by recording data or helping with
equipment. After completing training, volunteers selected areas for which they
Would be accountable for data collection during the study. They then worked
with the Off-site Team Leader and the Local Team Leader to establish a
monthly schedule with a built-in QA cycle. As volunteers collected their data,
thV completed data sheets were given to the Local Team Leader to copy and
file, with the originals going to the Off-site Team Leader for permanent project
records.
SUPPORT FOR VOLUNTEERS
Ensuring proper training and a positive atmosphere for the volunteers
helps ensure quality data collection. The most effective way to support volun-
An Approach to Improving Decision Making in Wetland Restoration and Creation
80
-------
teers is to keep them informed about the project and its progress so that they
know why they are doing what they are doing and why it is important. Volun-
teers will devote much time and energy to;a project when they feel that they
are part of the team. Creating opportunities'for local involvement in the study
of wetlands resources and their managemerit sends a clear message to the vol-
. unteers that public agencies are accessible ^nd local involvehnent is important.
Although they have less specialized knowledge and skill than their profession-
al counterparts, volunteers feel just as strongly about protecting the environ-
ment. Many citizens believe that natural resources are truly a part of the pub-
lic trust and as opportunities open up for cooperative ventures, they are ready
to work to protect resource values for all to share. The Trail's End projed
combined volunteers and scientists in an environmental partnership, and both
groups said, "It worked great!" >"
- Volunteers should receive feedback early in the project on how they are
doing. Immediate feedback about problems in data recording, questions or
technique, or concerns about procedures helps relieve anxiety about data col-
lection problems and helps prevent the possibility of lost data caused by im-
proper use of procedures. During training, it should be emphasized that volun-
teers are making important contributions to'the study and that they are free tc
telephone the team leaders about any concerns.
-: If possible, volunteers should be given: updates on the data they are col-
lecting, including any preliminary trends,:,patterns, or new information. Al-
though resource scientists usually wail: until data collection is complete to con-
duct analyses, volunteers appreciate preliminary observations, even though
limited, so that they can see the results of their workJVolunteers should not be
isolated after they have started work on a particular data collection task. They
need to contact each other and share information on experiences and topics ol
mutual interest. The Local Team Leader can play an important role in main-
taining this communication by sending brief newsletters or notes to the volun-
teers. Any publicity about the project (newspaper, television, newsletters, etc.)
should be brought to the attention of the volunteers before it becomes general
public knowledge. Weekly meetings during peak data collection times can
-also keep everyone informed. A debriefing session midway through the pro-_
ject and again near the end help meet this-need. At the" close of the project,
volunteers heed to know how project results will be used and why they are
•important in a larger context. The Off-Site Team Leader and the Principal In-
vestigator can help keep volunteers informed by sending them preprints ot
journal articles resulting from the project. - ~:
Some unstructured social time is important. Volunteers need to have some
free time with each other and with resource personnel, lunches and planned
break times give volunteers opportunities for extended conversations with stafi
on topics of common interest, in this case, wetlands. One major social event,
Volunteers and Natural Resource Monitoring
81
-------
such as a picnic away from the project, with family members included, can
add to the overall experience of;'everyone involved in the project. For the
Trail's End Project, a salmon barbecue at the home of the On-site Team Leader
provided an excellent setting for informal sharing. The research sponsor
needs to accomplish the project tasks, meet deadlines, consider costs, and
maintain the highest level of scientific standards, but personal exchange and a
shared awareness of the importance of natural systems at work do not have to
be sacrificed in the process. •
Opening research projects to local volunteers not only aids in data collec-
tion and other tasks, it heightens awareness in the community and among re-
source staff that protecting resources is everyone's job. It is important to re-
member that many people interested in becoming volunteers are working on
other resource, projects that may be of interest to the scientific staff. Their par-
ticipation can broaden the sensitivity of the resource specialists to the interre-
lated nature of community-based projects.
BENEFITS FOR VOLUNTEERS AND RESOURCE MANAGERS
Local citizens often believe that some distant person is in charge of their
resources. Typically, resource managers come to town, do what they do, and
leave. There may be some information in the local newspaper about their
work, but more likely there will not be unless major issues arise, such as
health, natural resource conservation, contested development, or infringement
on local government. It is unfortunate that resource monitoring and protection
often remain unnoticed. Inviting! interested local citizens to become part of a
study helps establish a line of communication with resource organizations. It
gives interested people an opportunity to contribute to the management
process. Most of all, it gives the volunteers a chance to join in the excitement
of conducting research and seeing the results used for making better resource
management decisions. In the case of Trail's End, it was satisfying and reward-
ing when the data the volunteers collected passed the EPA review process. The
scientists and volunteers had gained open communication, quality data, and
ongoing community involvement in the protection of local wetlands.
Involving local citizens at the research level creates a small, but informed
group of citizens who are sensitive to the protection of natural resources at a
more sophisticated level than the general public. Even though the project
ends, and the resource professionals go on to other problems, the new knowl-
edge and awareness, in this case! about wetlands and natural history, stays in
the community with the volunteers. Local volunteers who helped with the
Trail's End project have continued to collect data using the techniques learned
during the project. They -also added a bird banding element to the study be-
cause they wanted to know more about the migratory birds using the wetlands.
An Approach to Improving Decision Making in Wetland Restoration and Creation
82
-------
Individuals who become involved in a project of this type will begin to influ
ence their friends, coworkers, students, and local governments.
According to their own evaluations made at the close of the project, th<
individuals who participated in the Trail's End study found the experience an<
the learning more than worth the effort. The volunteers were doing what the
like to do best: being outdoors, protecting natural resources, and sharing tha
experience with others of similar interests. Volunteers left the project feelinj
that they had made a positive contribution, enjoyed the opportunity to worl
with dedicated researchers, become part of a local team, and contributed t<
wetlands protection. Many also renewed their connection with the earth. A
one Trail's End volunteer put it, "The wetland became my friend. I kept goin;
back to say 'hi' to the yellowlegs".
yd/unteers and Natural Resource Monitorir
83
-------
"|l:,i!illl'll1!"! •. I1!1 111!'!!! " 'Kill1 ,!'"
..'il'B 'I' "I1,""!1" " •.!!.. '"Ill1
USING EXISTING INFORMATION (2)
SETTING PRIORITIES/
SELECTING SITES (3)
MONITORING (4)
IMPROVING DESIGN
GUIDELINES
INFLUENCING FUTURE
DECISION - MAKING
-------
CHAPTER 5
Evaluating the Data and
Developing Performance
Criteria
The monitoring reports and other data collected for wetland projects an
keptjon file by many .state and federal, agencies? However, these reports an
rarely used. Often,the reports are reviewed^only to_document that the wet
lands were monitored as required (Quammen t9&6l.rr IrVthis chapter w(
demonstrate methods a manager can use to organize incoming data to tracl
the p^rdgfess of projects arid to develop criteria for the evaluation of future pro
jects. Specifically, we discuss ways to represent and evaluate the data, anc
ways to use the data to set performance criteria. Using the data to compare
past and present projects and natural wetlands can help managers set goals
anticipate future problems, and plan for the long-term success of wetland pro
jects. ........ -
SUGGESTED WAYS TO REPRESENT THE DATA COLLECTED
We use four-different graphs to display the monitoring data: performance
curves, summary or descriptive graphs, time series graphs, and characteriza
tion curves. These graphs differ in the amount(s) and type(s) of informatior
that can be obtained/and the intensity of data collection necessary. However
all can be used to compare projects and natural wetlands and, therefore, to se
criteria for the evaluation of future projects. In the following sections we de-
scribe the graphs and curves and present examples of how to create and use
them to set performance criteria. Most of the discussion is given to the perfor-
mance curves and the summary or descriptive graphs, because they were the
ones we used most often.
Chapter 5: Evaluating the Data and Developing Performance Criterh
87
-------
Performance Curves
The hypothetical performance curve described in Chapter 1 displays the
-hanges in a function in wetland projects over time as compared to similar
natural wetlands. As discussed in Chapter 3, the curve can be generated in
two different ways depending on how you sample. One approach is to follow
the development of similar aged projects by sampling the same projects and
natural wetlands over time (Figure 3-1 a). The other is to gather data from the
projects and natural wetlands at one time, but to document development by
simpling projects representing a range of ages (Figure 3-1 b). The latter is how
we geherated the performance curves from the results of our field studies (Fig-
ure 5-1). Because the age of the natural wetlands is not usually known, the
values of the mean and standard error for the measurement of function are
placed on the appropriate location on the y-axis.
Figure 5-1 shows that most of the projects had a lower level of percent or-
ganic matter than the natural wetlands, suggesting the pattern in the hypotheti-
cal examples (Figure 1-2). In reality the curve could take a variety of shapes.
If is possible that both recently constructed and mature wetland projects could
have a level of a particular function that is higher, lower, or the same as natur-.
a| wetlands. In addition/the pattern of development of projects could be ex-
pressed by a linear, quadratic, logistic,_or some other relationship (Figure 5-2).
Although we have a limited amount of information on the development of
projects, two things are beginning to be evident First the shape of the perfor-
rrjance curve will probably vary with wetland type and function. Second, we
believe that the pattern of development may be similar for the same wetland
type and function in different areas of the country. For example, we calculat-
ed a plant species diversity index for each of the created and natural wetlands
from the Connecticut, Florida, and Oregon Studies. Even though the means
for the created and natural wetlands varied between the states, in general, the
diversity of plants found on the created wetlands was initially greater than or
equal to that found on the natural wetlands (Figure 5-3).
There are many uses for performance curves, including evaluation and
comparison of projects sampled over time (Figure 3-1 a), and evaluation and
comparison of projects at one time (Figure 3-1 b). In addition, how frequently
to monitor the projects can be determined from the curves by noting yearly
changes in variables of interest If, for example, the percent organic matter in
the substrate did not appreciatively change from year to year, you might de-
cide that this variable need only be monitored in five or ten year cycles.
Specific management questions that can be answered include:
• What level of function is achievable for natural wetlands and projects
in a particular land use setting?
Chapter 5: Evaluating the Data and Developing Performance Criteria
88
! I '.
-------
20
18
16
14
i_ •
o
I
£ 12
.o
to
o 10
03
O
CD
J3._
c
CD
CD.
8
6
• = mean for 12 natural wetlands
I = ± 1 standard error
o = mean for 1 created wetland
I.
OO
O
O
2 3 4 5 6
Age of wetland (years)
Figure 5-1.
Performance curve generated using the mean percent organic matter in the upper
5 cm of soil from created and natural wetlands sampled in the Oregon Study. Mean
for the natural wetlands is the grand mean from 10 soil pits sampled at each of the 12
sites. Mean for the created wetlands is for 10 soil pits sampled at each of the 11
sites. Organic matter was measured as ash free dry weight.
Chapter 5: Evaluating the Data and Developing Performance Criteria
89
-------
:l It
-x-
•X-
•X-
ro uonounjio;
wetl
•9 3 '" 3
1 S.-3
I .& 1
:*
*
1
•X-.
-X-
•X-
•X-
uoiiounj ip sjnseayy
-x-
uoipunno
"c ^ '2 "
OJ O 'Q^ O
11 s-l
i||J
**• #~ O
f" O ~ u.
re _ .2 Q-
— ~& ~•
S "o « C-H
Q- > ni >« U
td gj t/i "2 "-3
_c — nJ e j I-T-
ui QJ
-------
w 0.7
lo.6
Q
0.5
0.5
I
.« 0.8
o
8.
| 0.7
0.6
0.5
•x0'9
i
C °'8 I
Q "" ™
CO
Q.
I 0.7.
'£
5
0.5
*
**
Connecticut Study
*
#
*
*
**
*
4 6
Age of wetland (years)
Florida Study
10
*
*
*
Legend
• = mean for natural wetlands
I = + 1 standard error .......
•*•= value for 1 created wetland
4 6
Age of wetland (years)
Oregon Study
10
4 6
Age of wetland (years)
10
Figure 5-3. Performance curves generated using plant diversity data from the Connecticut,
Florida, and Oregon Studies.
Chapter 5: Evaluating the Data and Developing Performance Criteria
91
-------
• Do the projects achieve the same level of function as natural
wetlands?
• How long does it take for projects to achieve the desired level of
function?
i ,;
• How can monitoring be timed so as to obtain the most reliable
information?
The WRP Approach is designed to evaluate wetland replacement in terms
of the wetland function actually replaced and the time required to reach the
level of function desired or possible. The information can be used a number of
ivays. For example, resource managers must include an evaluation of the po-
tential performance of a project when deciding whether replacement or
restoration is desirable. To accomplish this they need to distinguish between
projects that replace the function of natural wetlands and those that do not
(Figure 5-4, case D versus A, B, and C). They also need to know if a proposed
improvement to a site has potential for improving the current status of the re-
source (Figure 5-4, case B). Finally, they must distinguish those projects that
replace the function of natural wetlands in an acceptable time frame from
those that take prolonged time to mature (Figure 5-47 case A versus case C).
the time that it takes to replace functions is often ignored. Without: the incor-
poration of such knowledge into management schemes, we risk losing the or-
ganisms and processes associated with mature, natural wetlands because criti-
cal features of the mature systems can be lost before sufficient time has passed
for restoration and creation efforts to replace them.
Summary or Descriptive Graphs
Summary or descriptive graphs can be used to describe samples and iden-
tify outliers. There are many different types, but we found bar charts and box
and whisker plots especially useful. In our studies, bar charts were used to
compare a measure or indicator of function for a sample of created and natural
wetlands. For example, Figure 5-5a shows the percent of open water for each
• of the created and natural wetlands sampled in the Oregon Study. A graph
similar to a bar chart was used to compare the different weighted average
scores found in the Florida Study by ranking them in order along a line of pos-
sible weighted average scores (Figure 5-5b). Box and whisker plots (Figure 5-
5c) can also be used to describe a sample. The box outlines where the 25th
(lower), 50th (median), and 75th (upper) quartiles of the data are located. The
plot also gives an indication of the variability, symmetry or skewness of the
data, and the presence of outliers.
An Approach to Improving Decision Making in Wetland Restoration and Creation
92
-------
HYPOTHETICAL PERFORMANCE CURVES
O
LL
O -
tn
.CO
'CD'
legend . •
= Natural Wetlands,
i- .
= Restored Wetlands
Year of Monitoring . . .••• .
Figure 5-4. Hypothetical performance curves illustrating four different patterns of project de~ •
. , velopment that could be used in making management decisions. A population of
•natural wetlands is being compared to four populations of projects (in this case . .
restored wetlands) relative to a measure of wetland function. Population. A develops
' more rapidly than populations B,£, and D. A and C achieve the same level of func-
tion as the natural wetlands, while popu lation B exceeds the level 'of the natural
wetlands andI population D never achieves the level '.of the natural wetlands.-
Tirne Series GrapHs , ;
. Time series graphs are similar^to performance curves. Both display levels
of function versus time, but in the case of time s'eries graphs, "data points, are
not values for individual wetlands of different ages, but are observations from
one (or more) wetland(s) sampled over time (Figure 5-6). These types of
graphs work well with paired data (e.g.,'projects and natural wetlands in the
same watershed). The similarity of projects and natural wetlands, and whether
.or. hot levels of wetland function change with time, are two pieces of informa-
tion that can be ascertained from these graphs. In addition, jf water levels are
plotted versus time, the period when the water level is at, or above, the surface
can be used-to determine what portion of the site is jurisdictional wetland.
Characterization Curves ,
Characterization curves "are also referred to as frequency, distribution
curves or histograms (Figure 5-7). They are a type of bar graph with the verti-
cal axis representing frequency and the horizontal axis representing the vari-
able of interest, usually grouped into classes. For example, we might plot
number of wetlands versus an indicator or function of interest such as percent
organic matter found in soil. If you use number of wetlands for the y axis, then
Chapter 5: Evaluating the Data and Developing Performance Criteria
93
-------
I1 , ," ,' , ' ",1'IS .!'
'''IT** llil'llilnllltl
100
ro
! QJ
Q-
60
4°
"!!il
',!.!:: 20
1
I
Created witJzndj
Natural wetlands
0.99
0.79
1
S" 0.59
•5
I"9
o
0.19
-
-
-
• «— mixtmtm
(BUlflH)
-m«Sm
1— — roWmum
•
•
1
1
Created
wetlands
Natural
wetlands
Figure 5-5. Examples of summary or descriptive graphs, a) Bar graph of the percent of the site
• that was open'water on 11 created and 12 natural wetlands from the Oregon Study.
ONE BAR = ONE WETLAND b) Weighted average scores (Wentworth et al. 1988)
for the type of vegetation found on individual created (C) and natural (N) wetlands
from the Florida Study (adapted from Brown 1991). c) Box and whisker plot of the..
proportion of the plant community of created and natural wetlands from the Oregon
Study that was composed of exotic species.
An Approach to Improving Decision Making in Wetland Restoration and Creation
,v,; , ' : , ,< • , ' "... ,, ...94 '. , ' ' ' '. , , !
-------
30
20
10
o
CD
CD
Ground /
Surface *
CD
+->
CD
-10
-20
-30
-40
Created Wetland Natural Wetland
—e—
I A A I
9/88 10/88 11/88 2/89 3/89 4/89 5/89 6/89 .7/89 .8/89 9/89 10/89 11/8!
Time (months)
Figure 5-6. Monthly water levels (cm) for a pair of the created and natural wetlands from the
- Connecticut Study (adapted from Confer 1990).
—— Projects
--- Natural wetlands •
Low
Measure of function
High
Figure 5-7. Example of hypothetical characterization curves illustrating how natural wetlands
and projects might be compared. .
-• -- Ghapter-5: Evaluating the Data and Developing Performance Criteri,
95
-------
it is necessary to collect data from a large number of sites to generate these
curves. Therefore, they are not always a feasible option.
. N km , .'',i ' • ,, : • i,,. i"1, •!' , •' • 'I „ ij: • „ • , • , " „• k I '•li:i"ii
TECHNIQUES FOR DETERMINING DIFFERENCES IN SAMPLES
» The four graphic techniques are useful for displaying the monitoring data
ib that patterns can be discerned. However, managers often want to take the
data one step further and quantify differences observed when comparing pro-
jects to other projects or to natural wetlands. A number of statistical tests are
available that can be helpful. It is beyond the scope of this document to give a
detailed description of statistical tests and the assumptions for their use; how-
ever, we do provide brief descriptions of some tests and references for more
detailed information. Many introductory texts give detailed explanations of
|he tests and procedures described in this chapter. Snedecor and Cochran
(1980) is a good reference for statistical techniques such as comparing means,
variances, and slopes, and calculating confidence intervals. Devore and.Peck
(1986) is another good reference for basic statistical methods that gives de-
tailed examples of how to create different types of descriptive or summary
graphs (e.g., box and whisker plots).. Neter et al. (1990) is a good source for
information on regression techniques, and Krebs (1989), Sokal and Rphlf
(1981); and Ludwig and Reynolds (1988) give some specific methods for use
with ecological data. This list is by no means exhaustive, but provides refer-
ences for the tests if you need more information. In addition, statistical soft-
ware that will perform the statistical analyses is available.
The data used for the statistical tests must fulfill various assumptions (e.g.,
rjormal distribution, independent observations, equal variances) for the use of
the tests and the results to be valid. However, there are ways to manipulate
the data (e.g., transformations) so that they meet the assumptions, and there
are also alternative tests that can be used for different situations. Checking to
see if the assumptions of a given statistical procedure are met is a necessary
first step in data analysis and is described in the suggested references.
All four techniques—performance curves, summary or descriptive graphs,
time series graphs, and characterization curves—can be used to compare sam-
ples of wetland projects with samples of other projects or natural wetlands,
vVhile statistical tests can be used to determine whether there are significant
differences between samples. For example, you might be interested in
•whether the percent of organic matter in the substrate is different for a group of
projects of the same age that were sampled in 1980, and another group of pro-
jects of the same age that were sampled in 1990. Tests of hypothesis including
a Student's t-test or a nonparametric equivalent (e.g., Wilcoxon-Mann-Whitney
rank-sum test) can be used to determine whether the means of the two groups
aYe different.
"'An Approach to Improving Decision Making in Wetland Restoration and Creation
'• ": ' ' ' 96
-------
Another statistical tool that you could use is a confidence interval. Hy-
pothesis tests determine whether there is a difference between two samples
while confidence intervals provide a range of values for the difference in
means. For example, calculating a confidence interval based on projects sam-
pled in 1980 and separate projects sampled in 1990 would give you a range of
likely values for the difference in the population means between the two years.
This could be important to know. The level of organic matter in the substrate
is related to certain wetland functions such as water quality improvement. If
the design of the projects sampled in 1990 had been modified from the design
of the projects sampled in 1980 in an effort to accelerate thfe accumulation of
; soil organic matter, it would be helpful to quantify the difference between the
two samples. Higher levels of percent organic matter in the sample from 1990
could indicate that the changes in the design did accomplish the goal. To
make sure this was an actual improvement, other aspects of the projects
should be examined to confirm that important wetland functions were not af-
fected.
You could also compare the variability of two groups. Looking at summa-
ry graphs can give an indication of whether the variability of .two samples is
different. The range of the data can be determined from most graphs and gives
an indication of the variability. Box and whisker plots can also be used for in-
dicating variability. For example/in Figure 5-5cthe box and whiskers associ-
ated with the natural wetlands are longer than the box and whiskers associated
with the created wetlands. This indicates that there is more variability in the
data from the natural wetlands. You could also use a statistical test to deter-
mine if there was a difference. If certain'assumptibns were met,"tests such as
an F-test or a Levine's test could ascertain whether the variability of two sam-
ples was the same. However, in most cases looking at summary graphs is suf-
ficient, especially if the sample sizes are approximately equal.
When would it be relevant to determine whether the variability between
two samples was similar or different? One situation would be to determine
whether the sites within the samples were homogeneous. Another more spe-
cific example would be if you were monitoring water levels over time at a pair
of project and natural wetlands. Larger variability in water levels found at ei-
ther the project or natural wetland would be of interest due to the potential im-
plications for how the site could perform a specific function such as wildlife
habitat. ""
Multiple regression could be used to determine if there was a statistical
difference between the steepness of two lines. This would enable you to de-
termine whether the rate of change (or slope) on a performance curve from a
sample of projects was similar to the rate of change on a performance curve
from.another sample of projects (e.g., a sample from more;recent projects).
Certainly, it would be reassuring if the slope of the performance curve from the
Chapter 5: Evaluating the Data and Developing Performance Criteria
97
-------
sample of more recent projects approached the mean level of the natural wet-
lands faster than the slope from the earlier sample (Figure 5-4).
Statistical differences do not always imply meaningful biological differ-
ences. You will need to decide how much of a difference between two groups
is meaningful for your management or research needs, this points to the need
for confidence intervals in addition to tests of hypothesis. However, you may
simply want to rank the data to determine how specific wetlands compare
with each other, or use other purely descriptive techniques for comparing sam-
ples. Ultimately your professional judgment as a manager and an ecologist is
needed to interpret the tests and to decide if the results make sense. Results
that are difficult to interpret or do not make sense could indicate problems
with data collection, entry or analysis.
EVALUATING PROJECTS AND SETTING PERFORMANCE CRITERIA
The following example from the Oregon Study shows the process for set-
ting performance criteria based on vegetation cover data that was collected in
the field and then summarized in graphic displays. The wetlands studied were
all located in the Portland Metropolitan Area and consisted of 12 natural and
11 created ponds with a fringe of emergent vegetation (see Figure 3-5).
The percent cover of each plant by species was estimated in.0.1 or 1.0-m2
quadrats. The 0.1-m2 quadrats were used in wetlands where the vegetation
was short and relatively homogeneous. Forty quadrats were evenly spaced
along transects that were placed to represent the different vegetation commu-
nities present. Percent cover was used to estimate standing crop.
The mean percent cover per quadrat was calculated for each created wet-
land and then plotted.versus the age of the wetland (Figure 5-8). Since the age
of the natural wetlands was not known, the mean cover per quadrat and its
standard error were calculated for the sample of natural wetlands. Note that
most Of"the created wetlands Have values lower than the mean for the natural
wetlands. We could use this information to set performance criteria. A possi-
ble criterion based on our data would be that the mean cover of emergent veg-
etation on created wetlands during the first three years of development would
be less than that of similar natural wetlands.
:We also tested this more formally with a Student's t-test Using the mean
values from the individual wetlands as observations. The question of interest
was whether the mean cover per wetland was the same for the created and
natural wetlands. Th'e results of the Student's t-test indicated that the mean
cover per wetland was larger for the natural/wetlands (p=0.002). This result
makes sense because the created wetlands were younger than the natural wet-
lands. However, we would expect that with time the mean cover on the creat-
ed wetlands would increase to the level found on the natural wetlands. By re-
4/1 Approach to Improving Decision Making in Wetland Restoration and Creation
98
-------
160
140
CD -f-
I 120
D
cr
100
80
CD
Q.
»_
CD
o
o
•M
c
CD
O
L.
CD
Q.
§• 60
CD
40
20
Legend
• = mean for 12 natural wetlands
I = ± 1 standard error
•Jf = mean for 1 created wetland
Age of wetland (years)
Figure 5-8. Mean percent cover for created and natural wetlands from the Qregon Study plotted •
versus project age. This is the beginning of a performance curve for this set of pro-
jects. Mean for the natural wetlands is the grand mean from 40 quadrats sampled at
each of the 12 sites. Mean for the^created wetlands is for. 40 quadrats sampled at , ,
each~ofthe 11 sites. ~ ' '
Chapter 5: Evaluating the Datasand Developing Performance Criteria--
99
-------
M t
sampling the created wetlands in the future, we can get an indication of
whether the increase is occurring and at what rate.
The cover data also are displayed using a box and whisker plot. By exam-
ining Figure 5-9, you can see that the variability of the mean cover of the cre-
ated wetlands is larger than that of the natural wetlands. This would indicate
tfiat the sample of natural wetlands was more homogeneous in terms of cover
than the sample of created wetlands.
The diversity of the species found on the created and natural wetlands also
was calculated from the cover data. A performance curve of the data is shown
m Figure 5-10. In contrast to percent cover, the diversity of species on the cre-
ated wetlands tends to be greater than the mean diversity of species on the nat-
ural wetlands. Again, we would expect higher diversity because the created
wetlands are relatively new areas, and it is typical for a variety of species to In-
vaded we also would expect the diversity to decrease with time as cover and
competition for resources increased. The performance criterion set from this
data would be that the diversity of herbaceous vegetation on created wetlands -
during the first three years of development would be greater than or equal to
that on similar natural wetlands. Because the two samples were not normally •
distributed, we tested the difference between the diversity of the created and
natural wetlands with a"nonparametrfclest (Wilcoxon-Mann-Whitney) and:'
found that there was some evidence that they were different (p=0.09).
The presence of outliers can provide additional information. In Figure 5-,
10, there is one created wetland with lower diversity than the others. It is im-
portant to determine the reason for the low diversity and whether additional
investigation is needed. For example, another site visit could be necessary, or
rechecking the data sheets or field notes could be warranted. In the case of
our example, as a first step we checked our field notes. They documented that -
this site was receiving petroleum run-off from the parking lot of a city opera-
tions plant. We hypothesized that the run-off was affecting plant diversity (as
well as other attributes of the wetland) and that a site visit was needed to as-
cirtain what should be done to correct the problem.
A created wetland also could be an outlier and indicate a potentially posi-
tive situation, if it had an unusually low value of an undesirable function or a
high value of a desirable function. In these cases, we suggest that the wetland
be examined more closely to see'if there are any indications of ways to pro-
mote trie development of the desirable characteristic For example, suppose
the only mulched project .(see'Chapter 6 for definition) in a group of projects of
the same age, had a significantly higher percent organic matter in the substrate
than did other projects in the group. Then, a resource manager might want to
mulch future projects to determine if the mulching continued to enhance the
organic matter in the substrate.
An Approach to Improving Decision Making in Wetland Restoration and Creation
100
-------
250
200
co
•o
ca
Jf 150
-------
g °-9
•a
c
0.8- •
CO
03
-t 0.7
;i .ti
W
-
0'.5
*
*
*
*
Legend
• = mean for 12 natural wetlands
J| = ± 1 standard error
;(; = value for 1 created wetland
*
w « •—
Age of wetland (years)
Figure 5-10. Performance curve of plant diversity data from the Oregon Study.
the vegetation data also were used to compare the percent of species that
were common between the created and natural wetlands. We found that 41%
of all the species were found on both the created and natural sites. Forty one
percent of the species were unique to the created sites and 18% were unique
to the natural sites. However, the percent of species in common between the
created and natural wetlands was higher on a site-by-site basis (Figure 5-11).
The mean percent of species in common for created and natural wetlands was
68.1% (s.e.=3.1). A 95% confidence interval for the mean would be approxi-
mately 62-74%. This interval could be used-to-evaluate the means from future
samples of created and natural wetlands. As another option, we could deter-
mine whether future created wetlands fell approximately within the range of
values of our sample (45-81 %). We would be pleased if the created wetlands
fell into the upper end of the range, or above it, because this would indicate
that there were a greater number pf species in common between the created
|nd natural wetlands. The performance criterion set from this data would be
that 62-74% of the emergent or herbaceous species found on a created wet-
land during the first three years of development should also be found on simi-
lar natural wetlands. These types of analyses, and the associated criteria, also
could be used for cover data. In this case, the criterion might be—expect the
average overlap of species' cover to be between a and b for the created wet-
An Approach to Improving Decision Making in Wetland Restoration and Creadon
102
-------
100
CO
•a
—
I
(0
I
o
•o
o
»H-
o
CO
CO
CD
.'5
CD
a.
CO
CD
80
60
40
20
0
Created wetlands
Figure 5-11. Bar graph of the percent of species overlap between individual created and natural
' - wetlands from the Oregon Study (adapted from Gwin and Kentula 1990). ONE BAR
-•• = ONE CREATED WETLAND
lands during the first three years of development when compared to natural
wetlands. :' ' ; •''" '• ''
Finally, the vegetation data were used to compare the portion of species
that were wetland plants. A weighted average, calculated using the wetland
indicator .status of the plant (Reed 1988, appropriate volume PNW) and" its
cover value, was determined for each wetland (Wentw.orth et al.. 1988) (Figure
5-12). Note that the created and natural wetlands are interspersed along the
wetland/upland axis and generally fall between the values of 1.3 and 2.8. This
range indicates that there, is a good to high probability that the sites are .wet-
lands.. However, additional data regarding :soi Is land hydrology are necessary
to determine conclusively that some of the sites are wetlands. A performance
criterion set using this graph could be to expect values between approximately
1:0 and 3.0 for weighted averages calculated using data on herbaceous vege-
tation for both created arid similar.natural wetlands in the Portland Metropoli-
tan Area. ^Because the weighted average value, can be used to help determine
whether tfie vegetation parameter for delineation is met> it could be^ useful to
pick a specific cut-off value above which additional soils arid/or hydrotogy
data could be necessary for created wetlands. '_„..;;...-.. £...
As you can see from the discussion above, the vegetation cover data from
the Oregon Study were used in many ways. We were able to;display the data
Chapter 5: Evaluating the Data and Developing Performance Criteria
.703
-------
5.0
Extreme wetland
(100% obligate
hydrophytes)
Extreme upland
(100% obligate
upland species)
Wetland -«*-
-^ Upland
Figure 5-12. Weighted average scores (Wentworth etal. 1988) for the type of vegetation found on
individual created (C) and natural (N) wetlands from the Oregon Study.
tlMilllllUlll ' ! ' '",::' :„ i'1 'in'" I" IT i1 .'• -' i"."1 '' .''•! '" " ;"> '. '• 'v 'i1!'1, \,y " U"1'1'.1!1'1;1 '»•, J,1 •", mil 'I.:!!! iii1.,, /S1 i I'M, nil ' i „ ' li",1, ' ',.,,[ 'HiiP
jrj different graphs and to examine the variables found within the data set (e.g.,
percent cover, diversity, percent wetland species). We used the information
from the curves to answer questions such as: How does the species composi-
tion on the created sites comparei with that on the natural wetlands?, and, Is
tBe percent cover of vegetation similar for the created and natural wetlands?
Finally, we used the information from the curves to set performance criteria
that will aid in evaluating future created wetlands of the same type in an area.
The criteria that we assign will be useful to have in the future when we re-
turn to the Portland area to sample recently created wetlands. At that time, we
will be able to compare the data from the recently created wetlands with the
data from the original set of 11 created wetlands to determine if it takes less
time for the more recent group to become like natural wetlands. We will also
be able to determine if the recently created wetlands tend to fall within the cri-
teria we set based on the original sample. If they do not, we can hypothesize
as to why not. We will check for any new outliers, and use them to identify
what to do, or not do. For example, we can determine whether the changes
we made based on the outliers from the first group of created wetlands (e.g.,
the addition of mulch) were helpful. Sampling additional created wetlands
will enable us to develop more precise performance criteria and begin to make
An Approach to Improving Decision Making in Wetland Restoration and Creation
104
-------
decisions on whether the performance of the projects is adequately replacing
ecological function.
At the same time that we sample the recently created wetlands, we plan tc
resample the original 11 created wetlands. This will give us additional dat«
we can use to begin constructing the hypothetical performance curve illustrat-
ed in Figure 3-1 a, which in turn will help us to set performance criteria for fu
ture projects.
An Extension of the Example
A resource manager might have different states or regions to compare. Tc
illustrate how you could make such a comparison, the following example use:
data from studies that compared created and natural ponds with a fringe o
emergent marsh in Oregon, Connecticut (Confer 1990) and Florida (Browr
1991) (see Figures 3-5, 5-13, 5-14, respectively, for examples). Although, the
Connecticut Study used paired wetlands, it is included in this example for il-
lustrative purposes. Figure 5-3 shows performance curves of the plant species
diversity found on the wetlands sampled in the three states. Note that for eacr
of. the studies, although the mean levels of diversity for the natural wetland;
are different, the level of diversity for the created wetlands tends to be higher
The performance criterion developed using data from these curves would be tc
expect the level of plant diversity on projects during the first three to five yean
to be greater than or equal to the mean found on similar natural wetlands. /
resource manager whose data looked like Figure 5-3 would recognize any pro-
ject with a species diversity less than that of natural wetlands as a probable
outlier requiring further evaluation.
We do not want to imply that the greater diversity found on the createc
wetlands is necessarily a desirable or lasting phenomenon. .For example, the
diversity could have been due to weedy, opportunistic, or exotic species.
which points to the importance of evaluating the species composition. How-
ever, we consistently found the diversity to be greater in the newly createc
wetlands. Therefore, we would expect to find higher diversity in newly creat-
ed wetlands of the same type and age in different parts of the country.
Example of How to Use time Series Graphs
.... • Graphs of surface water levels; versus time, or of other variables where sea-
sonal changes might'6e expected, can show" trends over time and be used witr
: pairedwetlands or wetlands that.are,.grouped in some meaningful way (e.g.,
wetlands that-are hydrologically similar). Figure 5-6 shows a hydrograph frorr
the Connecticut Study (Confer 1990). As you can see, the highs and lows ir
water level occur at approximately the same time in the two wetlands. How-
ever" the created wetland appears to have water levels that are less variable
than the natural wetland. It also has a higher average water level than the nat-
Chapter 5: Evaluating the Data and Developing Performance Criterk
105
-------
:: I,1
=igure 5-13. Example of an emergent marsh in the Connecticut Study
* (S'f
Figure 5-14. Example of a pond with a fringe of emergent vegetation from the Florida Study.
^Approach to Improving Decision Making in Wetland Restoration and Creation
' 106
-------
ural wetland. How these two factors affect the fauna, vegetation, and othe
parameters associated with the wetland would be topics to pursue. Thes?
graphs can also help in establishing how much of the site is wetland. How
many days the water table is within a set distance from the surface and/or how
many days the water levels are at, or above, the surface are important factor:
in determining whether a site is a wetland. •' ;
Example of How to Use Characterization Curves
Figure 5-15 shows a histogram of the percent of organic matter found if
the upper 5 cm of soil in both created and natural wetlands sampled in the
Oregon and Florida Studies. We combined the data for this example because
there were hot enough data from each study separately to generate the curve
Since the data came from two studies, they are used for illustrative purpose:
only. You could use a display of this sort to determine the shape of the distrib
ution of the created and natural wetlands, to compare the levels of function fo
the created and natural wetlands, and to document the amount of overlap be
tween created and natural wetlands. After examining Figure 5-15, you migh
expect to see future created wetlands with soil organic matter values less thar
. or equal to that of natural wetlands. This could be tested directly with a hy
pothesis test. . ; •".
etlands
Number of
W
Created wetlands
Natural wetlands
_L
I
0-1 1-2 2-3 3-4 4-5
5-6 6-7 7-8 8-9 9-10 10-11 11-12 12-13 13-14 14-15 15-16 16-17 >17
Mean percent organic matter' . . .
Figure 5-15. Characterization curve of percent organic matter measured as ash free dry weight in
• . . the upper 5 cm of soil. Data are from the Oregon and Florida Studies. Error bars
have not been included since the data came from two differeht;studies. The example
is included for illustrative purposes only. Hypothetical curve is displayed in Figure.
5-7.
Chapter 5: Evaluating the Data and Developing Performance Criterii
107
-------
lUMMARY
In this chapter we have presented specific ways to graphically represent
;a'ta collected from monitoring wetlands. Four different graphs are suggested,
ach of which answers specific questions about projects and natural wetlands.
'he graphs are easy to create, and data can be added to them as they are col-
3cled. The more data collected and compiled, the more precise statements
laSed on the information will be.
In this chapter we also have suggested the use of graphic representations
if the data and statistical tests to develop performance criteria. The criteria
:an be used to evaluate the performance of projects based on the results of
»ast monitoring. Criteria for different regions can be compared and general
rends in the development and performance of wetland projects identified.
Finally, suggestions for improving wetland management can be made
>a§ed on this Information. Specifically, establishing performance criteria will
iid managers in decision making when defining objectives, anticipating future
jroblems, and planning for the long-term success of wetland projects. It also
vill enable managers to make knowledgable decisions as to when wetland
estoration, creation, or enhancement are viable options.
In Approach to improving Decision Making in Wetland Restoration and Creation
108
-------
(5)
(4)
(5)
USING EXISTING INFORATION (2)
V
SETTING PRIORITIES/
SELECTING SITES (3)
V
MONITORING (4)
EVALUATING THE DATA (5)
DEVELOPING PERFORMANCE
CRITERIA (5)
(6)
(6)
INFLUENCING FUTURE
DECISION MAKING
-------
' ".ill
•1 .'f
-------
CHAPTER 6
Improving Design
Guidelines
Information from natural and previously created or restored wetlands can
be used to evaluate the design of current projects and improve the design of
future projects. We present the results of several years of research in different
areas of the country to illustrate how to identify when the design of wetland
projects is or is not producing the intended results. This information can then
be used to plan and design projects that have an increased probability of per-
forming like natural wetlands.
We use two easily recognized wetland characteristics to present our case,
wetland type and vegetation. Other (characteristics of wetlands will be ad-
dressed, however not in great detail, because less information was collected in
the field on these characteristics, and the corresponding design information
from the project files was limited to nonexistent.
WETLAND TYPE
A significant finding of research conducted in Oregon (Kentula et al.
1992) and Wisconsin (Owen 1990) was that, although created wetlands tend-
ed to be located hrthe same county, river basin, or body of water as the asso-
ciated impacted wetlands, there were differences between the wetland types
impacted and those created. Therefore, local gains and losses of certain wet-
land types occurred. A similar trend has been observed on a national scale. A
recent report by the FWS (Dahl and Johnson 1991) states that, although gains
in some wetland types appear to offset some of the overall wetland losses that
occurred from the mid-1970s to the mid-1980s, many gains were simply con-
Chapter 6: Improving Design Guidelines
111
-------
rerslons between wetland types. Most significantly, gains occurred in non-
regetated unconsolidated bottom wetlands (i.e., ponds).
Determine if the Project is Typical of Wetlands in the Region
To use Oregon as an example, an analysis of the Section 404 permit
•ecord indicated that 23% of the wetlands created were ponds, but no natural
Ddnds were impacted. In addition, an examination of the NWI maps for the
Willamette Valley, Oregon, where most of the ponds were constructed, re-
galed that ponds were not a wetland type typical of the region. The only nat-
jfll ponds found were associated with major water courses, and, therefore,
4fre subject to yearly flooding. Typically, the created ponds were isolated hy-
irologically from rivers (Kentula et al. 1992).
Field research supported the analysis of the permit record, in that most
A/etlarid projects sampled in the Portland Metropolitan Area were ponds. The
structural characteristics that defined the projects as ponds included steep
banks sloping down to an expanse of open water with a fringe of wetland veg-
*tation at the water's edge. Because the structural characteristics that define
Zetland type also influence wetland function, wetlands of different types are
likely to perform different functions. Therefore, when making decisions about
vvhich wetland type to create or restore, one of the first considerations is, what
are the most important functions to replace. If natural wetlands in the local
landscape perform these functions, then the decision should be made to create
of restore wetlands of the same types. If there are compelling reasons for cre-
ating a type different from that which occurs naturally (i.e., need for flood de-
tgntion, sediment retention, or wildfowl habitat), the decision should be well
thought out and potential consequences anticipated. The decision should not
be based on what is the most convenient wetland type to create because of
available land or financial limitations. Furthermore, because we do not know
the ecological ramifications of replacing impacted wetlands with wetlands of
different types (Kentula et al. 1992), reason suggests that we err on the side of
caution and do our best to create types that occur naturally in the area. Un-
doubtedly, there are good reasons, geologically or hydrologically, as to why
the natural wetland types occur where they do.
Influence of Bank Slopes on Wetland Type
Data collected at created wetlands in Oregon (Gwin and Kentula 1990)
and Connecticut (Confer 1990) indicate that a large proportion were built with
steep slopes and consequently, only narrow fringes of hydrophytic vegetation
at the water's edge have become established. These created wetlands had no-
tably greater areas of open water than did similar natural wetlands sampled in
Oregon and Connecticut. The large area of open water and steep bank slopes
Of these projects resulted in ponds, rather than the palustrine emergent marsh-
i |
" 1' J ""Hill , , ' L_
An Approach to Improving Decision Making in Wetland Restoration and Creation
-------
es that were planned. Ten of 15 mitigation projects studied in Wisconsin also
resulted in ponds (Owen 1990).
Slopes and water depth influence the type and extent of wetland that will
result from wetland creation efforts (Owen 1990). For example, the steepness
of bank slopes leading into the wetland from surrounding areas influences the
extent of the vegetation community. Steep slopes provide less area at the ap-
propriate elevations and with appropriate hydrology for wetland vegetation to
become established. A narrow fringe of wetland vegetation is likely to occur
around a steep-sided pond, whereas, on a gentle slope wetland vegetation will
occupy a broad expanse.
Data collected to compare created with natural wetlands in Florida indi-
cated that, although the created wetlands did not.have greater proportions of
open water, they did have steeper bank slopes, greater basin depth, and conse-
quently, greater mean and maximum water depths than similar natural wet-
lands in the region (Brown 1991). The differences in slope between the natur-
al and created wetlands were probably the result of a;combination of
inadequate design and economic realities associated with the high value of
real estate. The created wetlands were built in residential or commercial de-
velopments (this is also true of projects sampled in the Oregon Study), often
tucked into corners, beside roadways or associated with storm water systems.
The lack of space and possible unwillingness of the landowner to commit larg-
er land areas (because of the high development value of the land) may have
contributed to the pattern observed where the amount of land needed for the-
wetland was decreased by increasing the slope of the banks. Gentle slopes re-
quire a. greater amount of land to achieve adequate basin depths and result in
larger transitional area surrounding each wetland (Brown 1991).
It is unlikely that ponds can replace the lost functions and values of wet-
lands that are filled (Owen 1990), and, as stated above, they often represent a
wetland type that does not exist naturally in the area. However, ponds are a
simple and inexpensive type of wetland to construct (McVoy 1988, Novitzki
1989), and are often favored over other types of wetlands because of potential
waterfowl habitat values (Gene Herb, Oregon Department of Fish and
Wildlife, Forest Grove, Oregon, personal, communication). Also, because
ponds can be tucked into small places, they are often the wetland type of
choice when there are constraints due to the amount or value1 of land available
for a project. When there is only a small piece of property available for the
project, it is often decided that a wetland of the same size as the property will
be constructed. Unless the ground surfaced at or very near the water table (in
this case, the area may already be a wetland!), steep bank slopes will be re-
quired to construct the desired area of wetland. Usually, this procedure will
result in the creation of ponds.
Chapter 6: Improving Design Guidelines
113
-------
A literature search found that most experts recommend that bank slopes
for created and restored wetlands range between 5:1 to 15:1 horizontal to ver-
tical (H:V)i However, recent research (Brown 1991, Gwin and Kentula 1990,
Owen 1990, Confer 1990) has indicated that bank slopes for most wetland
types should be constructed at or beyond the gentle end of this range (some-
where near 15:1 H:V or flatter) to make the wetland projects more similar to
natural wetlands; The slopes of ten of the twelve (83%) natural wetlands
measured in Oregon were flatter than 10:1, and the slopes of six of these ten
«7ere flatter than 20:1. Gentle slopes, that occupy a large expanse of the area
between the upland and any inundated area, allow development of a wide ex-
panse of wetland Vegetation rather than a narrow "ring around the pond" of
vegetation at the .water's edge. Figure 6-1 illustrates the difference in bank
slopes and Figure 6-2 illustrates the difference in the topographical profiles be-
tween created and natural wetlands in the Oregon Study.
The slopes of natural wetlands can be used as guides for contouring wet-
land projects. Therefore, to design a project to have topography similar to that
pf natural wetlands in the region, we suggest the following steps.
• ".. select a random sample of natural wetlands similar to the project type;
- .At each' natural wetland in the sample, establish transects to determine
the topography. A method for determining transect placement in
wetland areas along watercourses is presented by the Federal ICWD
(1989). The objective is to place the transects so that they are
representative of the site. Whatever method is used for determining
transect placement, the decision process should be documented;
';: .'"£• ;,_ . .- 'v:..; ••;' : • • ||,;; .'•• ••; ; " | ^ j ' • '. • ;'. :,"ll I;1';
." -. .Measure and record relative elevations at predetermined intervals
: , 'along each transect with transit and stadia rod. The measurement
interval along each transect will depend upon the size of the wetland,
the steepness of slopes, and microtopography; and
. • For each natural wetland sampled, determine the "zero" point from the
lowest elevation measured^ and adjust all elevations relative to zero.
Relationship Between Bank Slopes and Wetland Area
: bnceitie bank slopes for the desired wetland type have been determined,
tHe next step is to decide the amount of area required to construct the wetland.
In the following sections we will describe: 1) the factors to consider when de-
termining;how much land will be required for the wetland project; 2) the de-
Sign of the basin when a sufficient amount of land is available; and 3) the de-
sign of the basin when the amount of land is limited.
An Approach to Improving Decision Making in Wetland Restoration and Creation
114
II II
II
-------
a.
Figure 6-1. Pictures of typical natural (a), and created (b) wetlands in the Oregon Study.
Chapter 6: Improving Design Guidelines
115 .
-------
14 18 . 1B
Distance (m)
Figure 6-2. Topographical profiles for typical natural (a) and created (b) wetlands in the Oregon'
Study.
II 111 ; i : . ' i ,.. , . • , • i • •., . :i" - i ., v i. (.,.
Determinehowmuchland[will be required
As discussed earlier, a 1 -ha wetland will not fit into a 1 -ha piece of, land
Unless bank slopes are steep or water tables are at or near the soil surface.
ijierefore, if the project Is intended to be a type other than a pond or steep
sjded basin, sufficient land must be set aside to include gentle bank slopes
(which provide a transitional area between upland and wetland) and buffers.
, required slopes, in conjunction with the vertical distance from the precon-
ground' level to the water table.or water source, determine the
Tiuuniuf laneI required for the desired wetland area. The greater the distance
i ''thei'w'atertable, "the larger the project site must be to reach the water table
jf' ^gehtle" siojres. the pattern of the past has been to fit the project.to the
'$' ""'ivailable'site rather than to fit the site to the type of wetland desired. This pihi-
Sr:'tosoph'y"must be changed—the project site must be chosen to accommodate
the size and structural characteristics of the desired wetland type. Determina-
tion of projectsize requirements by this method may be more legally defensi-
ble than the ratios of wetland acreage to be created to offset wetland acreage
destroyed that are currently .used. Also, factors constraining the amount of
land available for the project must be considered in some situations. If the
project will be located within a highway meridian or residential or commercial
An Approach to Improving Decision Making in Wetland Restoration and Creation
-------
development, there will be constraints on the amount of Jand available, ant
the type of wetland that can be created.
Design when adequate land is available •
Information from our studies in Oregon, Connecticut/ and Florida, as wel
as other current research, can be used to make better informed decisions abou
the type of wetland and the features of the basin design for your project. First
determine how much land is available for the project. If the available land i:
adequate to construct the desired wetland type with the appropriate basir
characteristics, continue on with this section. If you will have only a smal
area, the next section describes the design of a basin when land is limited.
Decide what proportion of the available land will be inundated during the
driest time of the year, what proportion will be vegetated with hydrophytes
the width of bank slopes needed to provide a transitional area from upland tc
wetland, and the width of buffer areas. Use the proportions found within nat
ural wetlands in the area as your guide. Use the slopes of banks typical o
local natural wetlands as a guide for contouring the basin of the project. Foi
example, because the natural palustrine emergent wetlands sampled in Ore^
gon typically had bank slopes flatter than 10:1, we would recommend con-
touring projects of this type in Oregon with a variety of slopes 10:1 and flatter.
The next piece of information you will need is the vertical distance frorr
the water table or water source to the existing ground surface at the driest time
of the year. This distance can be determined with water wells or by digging
soil pits, and is the minimum depth the ground surface must be excavated tc
get water on the site if the project is to be: supported by groundwater. Then,
from your earlier decision on what proportion of the project ^should be inun-
dated during the wet (and dry) times of the year, you can decide where the
banks of the wetland should meet the water table. ; '
You now have the information from which to determine' how much land
will be required for the project. For example, if 15:1 (H:V) slopes are required,
the water table is 3-m below the ground surface, and you Have decided that
the banks should meet the water table at the bottom of the slope, a bank
length of approximately 45-m will be required to inundate the wetland at the
proper elevation via the required bank; slbpesrThis.distance must be added to
all sides of the inundated area extending out toward the surrounding uplands
and must be cpnsiderecj when determining the amount of land;required (Figure
.6-3).' ' '". •—; — ; ' : - ;;-.; ; ;——••-- -
Design when land area available is limited ir ^ "•"._"-','."."
If you must "fit" your project into a small area because, for example, it is
ancillary to a residential or commercial development, or must Be located near
a roadway or with in ~a highway meridian, you still have several 'options, the
- Chapter 6: Improving Design Guidelines
117 ' .-.•••...•
-------
Distance 1o ground water table
viaiuuu^ iw s.~-..- -- previous ground surrace
not to scale
Figure 6-3. Illustration of how to determine the amount of land needed for creating a wetland
!T given the bank slopes and the depth from the ground surface to the water table.
1 ' "' ' " " " ' " ' " ' I" " | ""
option most often used in the past was to construct a steep sided pond. How-
ever, we have the following recommendations that, if followed, should in-
crease the proportion of the site that will be vegetated.
Bank slopes should be as gradual as possible. This will increase the
amount of transitional area and the possibility for vegetation zonation along
the moisture gradient extending down from the upland edge, despite seasonal
and annual variability. Second, gradual slopes will help to stabilize the banks.
i?slSre'aretoo steep, the banks may erode or fail and slump into the wetland
causing that area of the wetland to be at an elevation higher than planned, and
consequently affecting the hydrology (Figure 6-4).
The wetland basin should be deep enough to attain the desired hydropen-
od for the intended vegetation community (Hollands 1990), but not so deep
that a pond will dominate the project. This means that the bottom of the wet-
land must be at an elevation where it will not be completely inundated, or
where it will be inundated only very shallowly, so that emergent vegetation
can persist.
VEGETATION ' , ".
Orie of the significant findings of the comparison of created and natural
Wetlands in Oregon (Gwin and Kentula 1990) and Florida (Brown 1991) was
that the composition of vegetation communities on the created wetlands was
not notably different from the composition of the communities occurring on
rjatural wetlands. Conspicuous differences did occur, however, between the
composition of the vegetation communities on the created wetlands and the
|f»ecies included on the planting lists for those wetlands (Gwin and Kentula
1990, Gwin etal. 1991).
An Approach to Improving Decision Making in Wetland Restoration and Creation
I w i „:,/, ;.., i. . • :••'.-, i:.,;." ji.9. ,:,.;
, • •
f :> J, | iiliji i f li ..... « ^i ...... , iilS, i|,i| i !
hil |f :> J,
-------
Figure 6-4. Erosion occurring on steep unvegetated banks at a created wetland sampled in the
' : Oregon Study. • • •
Example from the Oregon Study
. Comparisons between the vegetation that occurred on created freshwater
emergent wetlands in Oregon with the planting lists include^ in project plans
found veryfewVpecle'sTniommon (Gwin and Kentula:1990). The percentage
of all species found on a project that were included on the corresponding
planting list ranged from 0% to 7%. Therefore, between 93% to 100% of the
species that occurred on each project were volunteers. This finding suggests
that it may be unnecessary to plant freshwater emergent wetland projects.
However, before this inference could be made, we needed to determine if the
species that volunteered on the projects also occurred on natural wetlands, or
if the vegetation communities of the wetland projects consisted mostly of inap-
propriate species such asJnvasive exotics. ; _...,_. ;...
Therefore, we compared-the species that occurred on the, created wetlands
with the species that occurred on natural wetlands, and found that between
54% to 81% of the species were common to both groups. This suggests two
things: 1) the species included on the planting lists were inappropriate for ei-
ther the wetland types or the geographical area; and 2) planting lists should in-
clude the volunteer species because these species also occurred on natural
wetlands in the area.
Chapter 6: Improving Design Guidelines
119
-------
! !
it i
Example from the Florida Study
Comparisons similar to those in the Oregon Study were made between the
-reated wetlands and their plans in the Florida Study (Gwin et al. 1991), The
-esujts were similar to those found in Oregon. Vegetation communities at
three of the nine freshwater emergent wetland projects sampled were com-
posed completely of volunteer species. For the remaining projects, between
85% and 90% of the species found were volunteers. The species that oc-
curred on the projects were then compared with the species that occurred on
natural wetlands of the same type and size in the same area. This companson
indicated that the percentage of species on the created wetlands that also oc-
curred on natural wetlands ranged from 38% to 61 % (Figure 6-5a)
The analysis of vegetation was taken one step further in the Florida Study.
in addition to examining the species composition, we examined the relative
aBundance of each species. For those wetland projects that were planted the
percentage of the plant cover composed of species to be planted ranged from
6% to 33%. The majority of the plant cover, as well as the number of .species
on the project, was composed primarily of volunteer species. In addition, the
percentage of the plant cover on the created wetlands composed of species
that also occurred on the natural wetlands sampled ranged from 48% to 93%
(Gwin et al. 1991) (Figure 6-5b).
Guidelines for Revegetation of Wetland Projects
With the patterns described above in mind, we developed a generic ap-
proach to the revegetation of freshwater emergent wetland projects. The pri-
mary objective is to vegetate the projects with species appropriate to the de-
sired type of wetland (i.e., palustrine emergent marsh, riparian system,
shrub/scrub wetland, etc.) in any given region. The species that occur in nat-
ural wetlands and that have volunteered on previously created or restored wet-
lands form the ecological "blueprint" for revegetation. In addition, permit con-
ditions and specific project objectives are considered. To illustrate the
approach using the results of the Oregon Study, we developed a partial planti-
ng list for palustrine emergent marshes of the Willamette Valley which is dis-
cussed in the following sections. Structural components of the project (e.g.,
hydrology, slopes, soils, etc.) are assumed to be correct.
To Plant or Not to Plant?
Planting can be very costly, and in some cases may be unnecessary.
therefore, we begin with an analysis of whether or not to enhance or acceler-
ate revegetation by planting. The factors that contribute to the ability of a pro-
ject to revegetate with appropriate wetland species include:
"^ Approach to Improving Decision Making in Wetland Restoration and Creation
;: ;;, ;•; :; ' .' • 120 ,' ;".' ".' !l,
-------
b.
PERCENT OF SPECIES
100
80
PLANTED
101 H 102
106 H 204
VOLUNTEERS
H 103 I
B 205 I
ON NATURAL
104 H 105
208
PERCENT COVER
100
80
PLANTED
101 H 102
106 H 204
VOLUNTEERS
• H 103 •'• -.|
""'•'205' I
ON NATURAL
104- • 105
208
Figure 6-5. Comparison of the number (a) and the percent cover (b) of species found on created
wetlands in the Florida Study with what was listed in the project plans and what
occurred on similar natural wetlands.
Chapter 6: Improving Design Guidelines
121
-------
ii!! sum tm ms
I iSllli "1! it; "•'
. surrounding land uses and their contributions to the project in terms of
;,- .pollutants and undesirable seeds {lawns, crops); :
. , isolation of the entire project, or- a portion of it, from other wetlands
and appropriate seed sources (e.g., the centermost portion of a very
'.'. • . large project may require planting even if it is adjacent to another
wetland and appropriate seed source); •
":: - "••-.'- v •' -: -.-.• ." ..•:;•;<." •; •''•''"',€,'„:'•:^-"i-'\: • ::i li-i -i' •• •:" • ,/' '", 1-:';
-.;' vegetation strata, specifically whether herbaceous or woody species
are targeted to colonize the wetland (e.g., many herbaceous spec.es
. .volunteer and establish quite rapidly and, therefore, may not require
•,:';•• . '.planting; woody species often: take longer to establish and, therefore,
may require planting); '
• time of year that construction takes place; ;
' .' ' . . ,.•.-. ;;' . -. ' , ''.'." •< • , •„,! ' • •, • J> • '•',':•
' - • .hydrology, specifically timing and duration of inundation, water level
.fluctuations and flushing of the site; and.
• soils present on the site or any soil augmentation (such as topsail and
plant propagules taken from a destroyed wetland).
If the project is located downstream*' adjacent to, or nearby an existing
vegetated wetland, it is highly likely the project will have the ability to revege-
tate itself. As described above, our research indicates (Gwin and Kentula
1990, Gwin et al. 1991) that even when newly constructed projects were
planted, a high percentage of the species that occurred on the sites were vol-
unteers, arid.a large percentage of these species were the same as those that
occurred on local natural wetlands (Figure 6-5a). Therefore, although the time
required for a project to revegetate without planting may be longer than with
planting, if conditions are.correct, it may be appropriate to allow the project to
revegetate naturally. ' : : : _ - , .-
. tf you decide not to plant, revegetation .can be accelerated by mulching
the project with soils salvaged from a destroyed "donor" wetland, known as
salvaged marsh surface (SMS; Owen et a!, 1989). Mulching with SMS can ac-
celerate revegetation by providing seeds and other propagules; an organic sur-
face horizon, and soil microflora (Kruczynski 1990). It also makes the sub-
strate more conducive to rapid revegetation by .reducing the evaporation of
soil pore water, runoff/soil loss and erosion,.and surface compaction and
trusting (fhornburg;i977).: Although propagules:contained in the topsoiI re-
moved from the donor wetland should germinate on the project, a direct cor-
relation cannot be drawn between the vegetation that was present on the
An Approach to Improving Decision Making in Wetland Restoration and Creation
• '' ' 122 "
-------
donor wetland and the species that germinate from the seed bank contained in
the SMS. Studies have shown that the species that germinate from the seed
bank are often different from those thait were present on the donor wetland
(van der Valk and Davis 1976, Weinhold and van der Valk 1988). Species
generally referred to as "mudflat annuals" most commonly germinate during
the first few growing seasons after the wetland's constrution (Weinhold and
van der Valk 1988). However, in subsequent years, other species contained
within the seed bank should germinate and become part of the community.
Mulching has the potential to cause problems, however. Propagules from
species different from those that occurred on the donor wetland, or from unde-
sirable species may occur in the SMS. In addition, mulching may be unsuc-
cessful if the propagules were damaged during the excavation and stockpiling
processes.
If the project will be allowed to revegetate naturally, a monitoring program
should be instituted within a year after construction to ensure: that the project
does revegetate with desirable species. The monitoring program need not
consist of intensive sampling, but merely frequent routine checks to determine
if the project is becoming vegetated and what the dominant species are (Chap-
ter 4). If the project shows little sign of revegetating, if large areas of the site
are being affected by erosion, if important components of the desired vegeta-
tion community are missing, or if many of the species are undesirable, a
change in plan may be warranted and a planting scheme instituted.
Generating a Planting List '.••''
The vegetation community-desired on projects shoujd include those
species and communities that occur"on local/natural wetlands of the same
type. In addition, the community planted on the project should be "low main-
tenance" , i.e., it should be composed of plants that grow well and reproduce
at the given location in the particular climate with minimum care, and remain
free of serious disease or insect pests (Stark 1972). The following sections de-
scribe how to: 1) use the composition of the communities on natural wetlands
to generate a list of commonly occurring species; 2) determine which of the
commonly occurring species are commercially available; and 3) generate a
planting list from among the appropriate commercially available, commonly
occurring species. Finally, we give the reader additional guidelines to help en-
sure the success of the pi anting strategy. : .
What species commonly occur on wetlands in the area ?
••••-- Conduct a survey of the vegetation communities present on natural and
previously created and restored wetlands of the target type within the local
area. Identify all commonly occurring or dominant plants to genus and
species. In general, dominant species are those that contribute more to the
." •-- ,.;.-- Chapter 6: Improving Design Guidelines
123
-------
•* '':•:;,,
'ii j"
characteristics of the plant community than other species present, or that exert
I controlling influence on, or define the character of, the vegetation commum-
Iv Specifically, dominant species are those species in each vegetat.on stratum
(i e trees, shrubs or herbaceous layer) that, when ranked in decreasing order
of 'abundance and cumulatively totaled, exceed 50% of the total dominance
measure (i.e., percent cover) for that stratum, plus any additional plant spec.es
comprising 20% or more of the total dominance measure for the stratum (Fed-
eral ICWD 1 989). . .
1 From the vegetation survey/generate a list of the species that commonly
occur on the natural wetlands and existing projects. Th,s l.st reflects the
species that may be appropriate to plant on local projects.
"IE" ~ ....... , , ,,„..,,, ........ „ ....... . ....... ............. ..... ,.,.. ....... , ,, . ....... , [ . ...... ..
Which species are commercially available?
Check with local nurseries (especially those that specialize in native
species) to determine which species are commercially available. Compare this
list with the list of species that commonly occur on natural wetlands and exist-
ing projects to refine and narrow the list of species for planting. Then, deter-
mine the materials to be used for planting, (i.e., seeds, sprigs, culms, rhizomes,
potted seedlings, or woody cuttings) based on commercial availability. Also
consider the site conditions, how the plant materials are likely to be installed,
arid the goals of the project (elg., aesthetics, wildfowl forage, etc.). Seeds have
the least initial cost, but are likely to be lost from erosion or predation, and,
therefore may be more costly in the long run. Sprigs, culms, and seedlings
have a higher initial cost, but are more likely to establish successfully after
transplanting (Garbisch 1 986).
In addition to identifying which species are commercially available, it is
important to ensure that the nurseries propagate their own plants and do not
3 routinely remove them from natural wetlands in the area. However, it is also
important that the nurseries acquired their original stock locally (within the
study region). For example, until recently, many nurseries in the Pacific
Northwest acquired wetland species from nurseries in the Midwest (Rex van
Wormer, Independent Ecological Services, Olympia, Washington, personal
communication). Although a certain species may occur locally, if the plants
" were acquired from a nursery in a region with a different climate, they may not
II 'I'll »''"", ...... '."I" * . ........ ,. „ , ........ ,"• ,.,.,,„. ., ,.,.|» ., „ ..... ,' ,. a,,,' ». „ ,, " , ....... |, ,'i , , :,i .„.,„,, •• • • , , ;, ij .......... •
survive.
, , ,, , , , , ,, ..... , ,, , ,„ „,,,
Narrow the list of species to generate a planting list.
Use the appropriate Regional List of Species that Occur in Wetlands (Reed
1 988) to determine which commonly occurring and commercially available
'species are: " .................... ;" ......... " "" ' ' " ............. ' ....... " ...... " " ...... '
. • • ............. ........ i •• • ...........
• wetland species (obligate wetland, facultative wetland, facultative);
'[An Approach to Improving Decision Making in Wetland Restoration and Creation
-: ...... - ' • ' ' ' 'V J24 ....................
-------
endemic (i.e., native to the region); and
''.' « exotics that are not invasive and are part of the naltural community
(i.e., those exotics that typically occur on natural wetlands but do not
displace native species). .
Use this information to generate a sublist that consists of Qnly those plants
that are wetland species, endemic, and/or noninvasive exotics. Then, use a re-
gional flora to determine which species on this sublist are:
herbaceous, shrubs, trees;
weedy or opportunistic species;
• pioneering or early successional species; . ,,:.. . , ..
'•' common, hardy, or rare species; or
• . invasive (AVOID THESE).
The goals of the project will help determine which of these species should
be chosen for planting. For example, although mostly herbaceous species
should be chosen, for planting a pa lustrine emergent marsh, some shrubs
and/or trees might be chosen for planting along the tops of banks or to act as a
visual buffer. If soil •stabilization1!^ concern, "itmight be wise.to choose a few
species known for their rooting capabilities. , •
The above process will provide a list from which to choose species to
plant on the project. See Table 6-1 for an example of the types of information
this list might contain. The final list should include a minimum of plant
species adaptable to the various elevation zones within the project, diversifica-
tion will occur naturally. Garbisch (1986) recommends that one should:
Select herbaceous .species with potential value for fish and wildlife
- and with rapid substrate stabilization to help with initial
establishment;
Phase the establishment of woody species to follow that of the
herbaceous species and determination or stabiIization'of water levels;
Select species that are adaptable to a broad range of water depths. To
determine this, decide from the vegetation survey which species come
from wetter or dryer sites. In addition, inf6Tmatioh should be
Chapter 6: Improving Design Guidelines
125
-------
Table 6-1. Partial list from which to choose species for planting on wetland projects in the
Willamette Valley, Oregon. List was generated by following Steps 1-5 listed in the
section Generating a Planting List. Wetland indicator codes were taken from the re-
gional list of plant species that occur in wetlands (Reed 1988) and from consultation
with LaRea Johnston, Assistant Curator of the Oregon State University Herbarium.
Wetland indicator codes are: o—obligate wetland species; w—facultative wetland
species; f—facultative species; p—upland species; i—introduced species; n—native
species. The symbol + indicates the species is toward the high end of the category
(more frequently found in wetlands); - indicates the species is toward the low end of
the category (less frequently found in wetlands); and \ indicates the species is inter-
mediate within the category. Notes and characteristics of the species and common
names were derived from Hitchcock and Cronquist 1981, Steward et al. 1963, and
• Newling and Landin 1985).
SCIENTIFIC NAME
COMMON NAME HABITAT
:' '• Jy, ,['.,-<• , ;!•![ CODE
NOTES/CHARACTERISTICS
Emergent Forbs:
Jjlisma plantago-aquatica Water plantain
o\n
Iris missouriensis
Western iris, , w+n
Rocky Mountain iris
Submergent or Floating Forbs:
Lemna minor Duckweed
Oenanthe sarmentosa Water parsley
o\n
o\n
inili , , ' ' •, ' " • :i
Forbs;
Geum macrophyllum Large-leaved avens w+n
Lysichitumamericanum Skunk cabbage o\n
Crasslike:
Carex aperta
Carex obnupta
Ferns:
Athyrium filix-femJna
':'!"!!'! ! j; "
-1!«'i. ;.,*
Shrubs & Trees;
;-"'i:"' Corylus cornuta
Fraxinus latifolia
Columbia sedge w\n
Slough sedge
Lady fern
Filbert, Hazelnut
Oregon Ash
o\n
-f\n
p\n
w\n
Widespread in North America; shallow
.:. water & wet areas; good food source for
::; wildlife. ' ' '
Pale to deep blue flowers; wet meadows and
streambanks, but can tolerate dry summer
condtions; B.C. to Calif & east to S. Dakota!
• i j
Temperate & subtropic freshwater lakes,
ponds & slow moving water; good water-
fowl forage.
••••"•• Widespread along Pacific Coast, Alaska to
-'"" Calif.; still or sluggish water.
Si1!,".!,' '.if i il'i. ' ! •• ;l'" '' ii: . . ;• ' '" ' , '/ill )
Widespread on wet ground from near sea-
level to subalpine; Alaska to Baja Calif, east
to Rocky Mtns.
. Mephitic; swamps & bogs from Alaska to
Calif. & east to Idaho; easily propagated by
si division of underground stems.
';'':::; ' '•"' "I •• ' ; : '• •'• ••"• !' ' "i : ' • ;
Wet lowlands, esp. floodplains;
B.C. to NW Oreeon & east to Idaho
•"-II'.T;,,i"*."- "WWT1"!.—'« ° - ' • . — ,
& NW Montana.
Wet ground or standing water; Cascades to
coast, B.C. to Calif; soil stabilizer & wildlife
('"• forage.
,
Very common, lowland to montane, circunri-
boreal; woods, meadows and swamps; can
I be a pest. '
Widespread at low elevation on well
drained soil, B.C. to Calif & east to Idaho.
Deep fertile moist soil, esp. streambanks;
B.C., west Cascades to Sierran & coastal
! caiif. "! !'.. ' _";.., ' . ;.
An Approach to Improving Decision Making in Wetland Restoration and Creation
: " "• • " ' " • ! ' " • 126
-------
available from the nursery. Most nurseries wiI
specifications for planting (e:g., In damp soils, d
be happy to provide
i of.water, etc.);
depth
Avoid choosing only those species that are
'expected to use the site. Muskrat "eat-outs" of"
created wetlands sampled in the Connecticut
resulting in a complete loss of vegetation at thes
at one site recovered and became more div
vegetation at the other site did not recover ir
part to excessive flooding, the site remained
next growing season; and
foraged by wildlife
'ypha occurred at two
Study .(Confer 1990),
; sites. The vegetation
erse. However, the
im mediately, and due in
regetated through the
urvi
Avoid committing significant areas of the site
questionable potential for successful establishing
to species that have
nt. ;
and substrates to the
OTHER IMPORTANT STRUCTURAL CHARACTERISTICS
Although we recognize the importance of hydrology
success of each wetland project, information from our research that can be ap-
plied to project design is limited. Therefore, our discu
-------
Table 6-2. The hydrology
metropolitan
Engineers (CCJE)
Environmenta
ay planned for created wetlands studied in the Portland, Oregon
Irea in 1987. Information was taken from the U.S. Army Corps of
and the Oregon Division of State Lands permit files. EPA = U.S.
Protection Agency.
HYDROLOGY INTENDED AT CREATED WETLAND
Letter fro in
rriaintaineld
adequate
EPA to COE states that a hydrauljc connection must be
between the project site and the adjacent creek to maintain
stream flow for fisheries.
C2-TI
O3-NS
Design p
basin in
an shows a pipe leading into the created wetland from boat
Columbia River.
the
Special
dugwetla
Condition 8 of Attachment A to Permit states:"..'. connect newly
nd into the existing stream." .
O4-MHP Lake to r
corners.
, . . . . i,,,, J.,1 ff!'\ , ! • "I1,, BC'lij! /;;if .Sit
sceive water'from two streams entering at its NW and SW
•he streams drain a 572.1 acre watershed.
ill i
C5-MG
Well wabk is to be supplied to the lake during seasonal low stream flow
to mainta n the water depth at agreed upon levels.
Drawings
street.
show a culvert leading into the wetland from under nearby
Excavation to the level of an adjacent stream area subject to stream
overflows and possible periods of standing water. '
C6-3I Existing creek channel to be rerouted through created wetlands. Stream
flow estimated as about 4 cubic feet per.second.
I!" '!,,"! ! ;!
:* 'i Ill, . , i! . 'ii,
•(• iiiiilS ' ' !• 1
!p '"''!:!!• ' i ' 1
if! ,'i'i'1''1"' ' 1 ,•" i1
ui '•! '•
* :•'. ' : ' : •
,!n!: ii?1!!1!!! , , ii ' '
ii t i: !,•: i ••'
.us "M; • : 'i! :
ii :: ii
C7-SML New stre
supply w
overflow
C8-BSP Drawing
The over
•' i ' :* • " n .• ""-• ' "'slough ar
•I!'!1! , : , . ' i II i ,• , : - • , •
1,,,'i" • ' "Roof wa
.k! i. i " '.•'•:•>'•"'
'<;< i i. i i -"... . i," •• • ',
Surface v
. . ^ .. .... „... _ |he basin
: , ;::,: ,.;, •„ . Text sta!
'! , , .I'll1 ; ,,,31111 . „ ,;: '„. , M.;'!,, ,"! ' i , '
{;> . •, • ',;' {.j' "|1|,(;:;: ~~' • ' • , ',i;; sub-surfa
v ' " i *!! '..in • •. ..'".i. '••:.
An Approach to Improving
,., ; ,,;;
am channels to be excavated to increase stream length and
ater to project. Existing stream channels to be maintained as
channels.
shows overflow slough connecting pond with nearby creek.
.... • i . ' • , . i .,
flow channel is to be created between the existing overflow
d the SE corner of the project site.
er" will be discharged from two buildings into the pond.
/aters from the surrounding developments to be discharged into
through diffuser pipes. . . ,
es that "there may always be a slight freshwater flow from
ce seepage". '
Decision Making in Wetland Restoration and Creation
128
I i i i I'M i ill I I I i i I «' i'1'll
-------
signing the project to relate to the hydrology of the site. For example, the con-
struction of hydrologically isolated ponds in areas such as Oregon where nat-
ural ponds are usually connected to a body of water (Kerituta:et al. 1992), will
cause the ponds, although structurally similar to their natural counterparts, to
function differently hydrologically. For the created ponds to function in a
manner similar to that of natural ponds in this area, they must be hydrological-
ly connected to streams or rivers. In addition, creation of a structurally similar
wetland on substrate different from the substrate of natural wetlands may not
facilitate similar hydrology because of differences in permeability (O'Brien
1986).
A .substantial amount of hydrological information can be obtained from
local natural wetlands with a modest Investment in supplies; and equipment.
Water levels can be recorded continuously with water level recorders, or by
reading a staff gauge during periodic site visits. With water level data, most
hydrologic variables can be determined—hydroperiod, flooding frequency and
duration, and water depths (Mitsch and Gosselink 1986). The development of
water budgets for natural wetlands may further increase the probability.of suc^
cessfully creating or restoring a wetland, because the water budget provides a
design for the hydrologic characteristics (Novitzki 1982), However, because a
water budget is based on inflows equalling outflows, great care must be taken
to ensure all components of the equation are accurately measured, and poten-
tial errors and their causes are estimated (Winter 1981T. 'Atypical water bud-
get equation is: " P + OF + SWI + GWI = ET + SWO + R, where
P = precipitation on the wetland in inches or centimeters,
OF = overland flow into the wetland, .
SWI = stream flow entering the wetland,
GWI = groundwater inflow to the wetland,
ET = evapotranspirative losses from the wetland,
SWO = stream flow leaving the wetland, and
R- ~" =; re£harge from the wetland to groundwater. -
Other hydrological data to collect on local, natural wetlands include:
flow conditions (i.e., whether water is.flowing over the site or whether it is
.mostly stagnant, and whether it flows quickly or slowly); wh,ether~the flow of
water is channelized or sheet flow; whether the ground was inundated or satu-
rated at some distance from the surface or at the surface; thfe proportion of the
wetland that is covered with open water; seasonal water level .fluctuations; and
locations and types of water inflows and outflows. In some regions certain of
these data may already exist and be used in project design. For example,
---•—— chapter 6: Improving Design Guidelines
129 '
-------
"I1 1
Golet et al. (in press) document normal water lever fluctuations in red maple
swamps' in Rhode Island, and Kantrud et al. (1989) describe the hydroiogic
regime of prairie basin wetlands in the Dakotas. It is very important that the
hydrological characteristics of the project are documented in the design and
Construction plans for determinations of compliance and so that successful
projects may be used as models for future hydrological design.
Soils/Substrates
.•;::.,; Data collected in the Oregon Study showed that most created wetlands
had significantly lower soil organic matter than did natural wetlands., The av-
erage-percent organic matter in soils of projects (5.49% at 5-cm depth, S.E. =
1.05%). was significantly .lower than that of soils of similar natural wetlands at
depths of 5-cm (10.13%, S.E. = 1.67%), 15-cm, and 20-cm (p=0.002, p=0.02,
p=0.02 respectively). Due to the young age of these projects, the lower or-
ganic matter was expected. What was unexpected was that one created wet-
land had organic matter much higher than all the other created wetlands and
the mean for the natural wetlands. Further examination of this project could
lead to insights into how to accelerate the accumulation of organic matter on
other projects, concurrently increasing wetland functions related to soil organ-
ic matter content.
Soil organic matter is an important potential source of available nitrogen
(Langis et al. 1991). In addition, soil organic matter stores nutrients and pro-
vides organic substrates for bacteria involved in nitrogen fixation, denitrifica-
tion, and the sulfur cycle (PERL 1990). The lower soil organic matter of the
created wetlands suggests that these soils have less energy for soil microbes to
recycle and fix nitrogen, and because of the low nitrogen inputs, plant growth
will be limited (Zedler and Langis 1991). Conversely, in systems with high ni-
trogen inputs, the low organic matter in created wetlands might limit the sys-
tem's ability to process nitrogen through denitrification because of low carbon
availability (Faulkner and Richardson 1991), and thus constrain water quality
improvement values. Over time, we would expect the organic matter of soils
of wetland projects to increase. However, because we as yet have no data on
how long it will take organic soils to develop, enhancing the percentage of or-
ganic matter may be the best way to .accelerate the development and facilitate
the development of related functions. .
Augmenting the substrate of wetland projects with SMS (Owen et al. 1989)
from a donor wetland will make the substrate more similar to that of natural
vvetlands, and provide a possible source of appropriate wetland plant propag-
jles. In addition, because organic soils have a higher capacity for water reten-
:ion and an increased proportion of this water is available for plant growth, the
Qrobability of wetland vegetation establishment is increased. Organic soils
also have higher cation exchange rates and consequently a higher buffering
An Approach to improving Decision Making in Wetland Restoration and Creation
130
II"! t '•:"!! '
:.»j - •
-------
capacity than do mineral soils (Brady 1974). Because organic matter has a
high capacity to complex or adsorb metals and organics, the amount of organ-
ic matter in the substrate can influence the wetland's potential for pollutant re-
tention.
The contours of the project should be graded before the destruction of the
natural marsh so that the SMS can be transferred directly. In any case, the
SMS should not be stockpiled longer than 30 days because of possible oxida-
tion of the soil, possible release of metals that may be toxic to seedlings, and
possible loss of viability of some seeds (Brooks 1990). Whefi transferring the
SMS to the project, it should be spread over the substrate carefully, with mini-
mal handling, overturning or trampling.. If SMS is not available, there may be
readily available sources of waste organic matter to augment mineral soils,
such as municipal leaf/grass compost, composted livestock bedding and ma-
nure (although seeds of aggressive weedy species may be present), and food
processing wastes. ' j
Although the role of mulching or augmenting the organic matter content of
soils is not yet clearly understood, we recommend augmenting the soils of pro^-
jects to make the organic matter content more like that of natural wetlands.
Further research will then provide insight as to whether or riot augmentation
accelerates the development of these projects. ; '
SUMMARY
Will changes in the design of well and projects cause them to develop
faster and become more like natural wetlands? Will they be a better "fit" in the
landscape? Our interpretation of the results from field studies so far, indicates
this may be true. We suggest that better wetlands can be designed by model-
ing projects on local natural wetlands and on what Was learned from earlier
projects. We contend that this will lead to ecologically based performance
criteria for wetland restoration and creation that will, in turn, lead to better
management and protection of the resource. j
Looking to the future, we intend to continue building the knowledge base
on wetland restoration and creation through the application, testing, and eval-
uation of the concepts presented in this document. The research to be imple-
mented by EPA's WRP in the coming years williattempt to fill some of the gaps
we have identified in the course of our studies to date. As stated earlier, there
is a paucity of long-term data on the development of wetland projects. The
projects we have described will soon be five years older. It will be a priority
for us to repeat at least one of the three studies (i.e./ Connecticut, Florida, and
Oregon) to generate the next part of the performance curves. In this way we
can further document the development of these freshwater wetland projects.
' We have reported on the most common type of mitigation project nation-
ally, a pond with a fringe of emergent marsh. Although they are very com-
"""'" Chapter 6: Improving Design Guidelines
131
-------
mon, they are not the only type of wetlands being restored and created, or the
only type being studied. Table 6-3 summarizes the findings from recent stud-
ies of groups of wetland projects. We are looking forward to applying our Ap-
proach to projects involving other wetland types to begin documenting their
performance and to expand the scope of our Approach. Specifically, we will
begin focusing on the restoration of riparian systems in the arid West in the
near future.
We maintain that consideration of ecological setting is important to evalu-
ate and understand the functions of natural wetlands and the performance of
projects. Determining the effects of different land uses on wetland function
will be a major theme of our upcoming research. Such information is neces-
sary for both the protection of the wetland resource and the success of restora.
tion and creation projects. With knowledge of the effects of surrounding land
uses, appropriate management strategies can be employed to protect key wet-
lands, e.g., the use of buffers. In addition, knowing how present arid projected
development of an area will affect wetland function can influence decisions
on how to prioritize sites so that projects maximize ecological benefits.
> Fundamentally, as we plan and implement new studies we will continue
to treat existing projects as experiments in progress and promote the idea that
we all must
"...learn by going where we need to go.'.? (Roethke, 1961).
!i. -
An Approach to Improving Decision Making in Wetland Restoration and Creation
I, „.,.;: •. 132
-------
FINDINGS
a)
U.S
OS"
WETLAND
TYPES
z
o
5
o
|
£
5
Success was related to adequacy of planning,
design, implementation and follow-through.
*~
c
to
saltwater, brackish,
freshwater
CO
8
CO
o
CD
cn
"ro
"£
(D
•1
CO
CJ
Features of sites that did not meet agreed-upon
criteria in the permit plan, or the permit criteria did
not address habitat trade offs adequately.
^™
0 0
salt marsh; mangro
habitat and freshwa
o
*o 3
i u
03 TO
C « *C
Sag
o
en
en
°
Only 4 projects met all stated permit goals. 1 6 of
the failed or incomplete projects were correctable,
but 6 could not succeed under any circumstances,
and 14 projects required more study to determine
the feasibility of corrective actions.
*
I
M
03
forested arid non-fo
freshwater; wetland
S
U_
P
en
m
A high rate of noncompliance was found. Only 4 of
the 63 projects reviewed were in full compliance
with permit conditions. The ecological success of
sites built was 1 2% for freshwater systems and
45% for tidal systems.
CD
-o "to
Vt "O >-
• i S «?•-•-•
freshwater herbace
forested wetlands,
herbaceous and ma
wetlands
1
-•s-
e -
P en
B — en
&!c
0 f .9
.-is •§.
g > 0)
in ui m
None of the projects were constructed as permitted
or planned. A cumulative loss of 29% of the area to
be created occurred, and vegetation occurring on the
projects consisted primarily of volunteer species.
CO -
*-, CO
palustrine erriergen
open water system
P
c
o
£
o
•a
t
£
CO
a
1
•§
"o
- .£ en _
jsen
O —
Correlations between area required and area as-built
could not be made for 6 of the created wetlands
.because of inadequate information in the project
files. Vegetation occurring on the created wetlands
consisted primarily of volunteer species.
en
•" co
palustrine ernergen
open water:System
}
•o
•c
o
E
E
~
en
-'CO
«
c
--•£••-•-•- •-•"
U _..
Eleven of the 18 sites resulted in a-net loss of area.
Four sites resulted in wetland types partiaiiy or
entirely the same as those lost. Two have a good
chance of becoming the type that was lost. Nine
have incorrect physical conditions. Three were not
constructed. . .
CO i
•W?
freshwater '.wetlanc
mostly ponds
•i
c
§
o • • • • i - • -
m
c
-P
Vegetation characteristics were highly variable, but
properly planned, constructed and maintained sites
provided viable wildlife habftat.
CN •
01
coastal marshes
(0
&
cn
en
i
QJ
.Q
O
1
J.
**—
o
t/1
CL
I
"8
Chapter 6: Improving Design Guidelines
133
-------
til' !!
-------
REFERENCES
Abbruzzese, B., A.B. Allen, S. Henderson, and M.E. Kentula. ;1988. Selecting
sites for comparison with created wetlands, p. 291-297. In C.D.A. Rubec and
R.P. Overend (Comp.), Proceedings of Symposium '87—Wetlands/Peatlands.
Environment Canada, Ottawa, Ontario, Canada.
Adamus, P.R. and K. Brandt. 1990. Impacts on Quality of Inland Wetlands of
the United States: A Survey of Indicators, Techniques, and Applications of
Community Level Biomonitoring Data. EPA/600/3-90/073. U.S. Environmen-
tal Protection Agency, Environmental Research Laboratory, Corvallis, OR. .
Anderson, J.R., E.E. Hardy, J.T. Roach, and R.E. Witmer. 1976. A Land Use
and Land Cover Classification System for Use with Remote Sensor Data. U.S.
Department of Interior, Geological Survey, Professional Paper 964. Washing-
ton, DC
Bedford, B.L. and E.M. Preston. 1988. Developing the scientific basis for as-
sessing cumulative effects of wetland loss and degradation on landscape func-
tions: Status, perspectives, and prospects. Environmental Management
l2(5):751-772. ~"
Berger, J.J. (Ed.) 1989. Environmental Restoration: Science and Strategies for
Restoring±he Earth. Island P^e^s, Washington, DC. ;
Brady, N.C. 1974. The Nature and Properties of Soils. 8thjed. MacMiilan
Publishing Company, Inc., New York, NY. !
Brooks, R.P. 1990. Wetland and water body restoration and; creation associ-
ated with mining, p. 529-548. In J.A. Kusler and M.E. Kentula (Eds.), Wetland
Creation and Restoration: The Status of the Science. Island Press, Washington
DC.
Brooks, R.P. and R.M. Hughes. 1988. Guidelfneslor assessing the biotic
communities of freshwater wetlands, p. 276-280. In J.A. Kusler, M.L. Quam-
me'n, and G. Brooks (Eds.), Proceedings of the National Wetland Symposium:
Mitigation of Impacts and Losses. Association of State Wetland Managers,
Berne, NY. "" ^":;~•"'""""• •' ••-->• — •••• - __;__ _ ', , '
Brooks, R.P., E.D. Bellis, C.S. Keener, M.J. Croonquist, and D.E. Arnold. 1991.
A methodology for biological monitoring of cumulative impacts on wetland,
stream, and riparian components of watersheds, p. 387-398. In J.A. Kusler and
135
-------
S. Daly (Eds.), Proceedings of the International Symposium: Wetlands and
River Corridor Management. Association of State Wetland Managers, Berne,
NY- .'.'...; ;;,;.;;,...„;:•' ;:;.;',:;.;, '.;;;;, ./::;]:.;.
Brower,"j.E. and J.H. Zar. 1984! Field and Laboratory Method's for General
Ecology. 2ndled. William C Brown Co., Dubuque, lA.
.. • ' - •' : ' | • ' :["
Brown, M.T. 1991. Evaluating Constructed Wetlands Through Comparisons
with Natural Wetlands. EPA/6p6/3-91 /058. U.S. Environmental Protection
Agency, Environmental Research Laboratory, Corvallis, OR.
Brown, MX,"}. Schaefer, and K. Brandt. 1990* Buffer Zones for Water, Wet-
lands, and Wildlife in East Central Florida. Publication Number 89-07 and
Florida Agricultural Experiment Station journal Series Number T-00061. Cen-
ter for Wetlands, University of Florida, Gainesville, FL.
Cahoon, D.R. and R.E. Turner. 1989. Accretion and canal impacts in a rapid-
ly subsiding wetlands. II. Feldspar marker fibrizori technique. Estuaries
12(4):260-268.
•• • '• • • • • • " • '^; • •••"'^••- '"!':;" i";" •' '•, " '• •t '::; l
Cairns, J., Jr. (Ed.). 1988. Rehabilitating Damaged Ecosystems, Volumes I and
II. CRC Press, Boca Raton, FL.
/^ Qiabreck, R.H. 1988. Coastal Marshes: Ecology and Wildlife Management
University of Minnesota Press, Minneapolis, MN.
Confer, S.R. 1990. Comparison of Created and Natural Freshwater Palustrine-
Emergent Wetlands in Connecticut. M.A. Thesis, Department of Botany, Con-
necticut College, New London, CT.
Confer, S.R. and W.A. Niering. In press. Comparison of created and natural
freshwater emergent wetlands in Connecticut. Wetlands Ecology and Man-
agement.
The Conservation Foundation. 1988. Protecting America's Wetlands: An Ac-
tion Agenda. The Final Report of the National Wetlands Policy For.urn. Wash-
ington, DC.
(Jiwardin, L.M., V. Carter, F.d Golet, and ET. LaRoe. j 979. Classification of
Wetlands and Deepwater Habitats of the ''United States. FWS/bBS-79/3ll
USS. Fish and Wildlife Service, Washington, DC.
136
-------
Crabtree, A., E. Day, A. Carlo, and G. Stevens. 1990. Evaluation of Wetland
Mitigation Measures, Final Report: Volume I. U.S. Department of Transporta-
tion, Federal Highway Administration Report Number FHWA-RD-90-083.
Washington, DC. •
Crewz, D.W. 1990. Habitat-Mitigation Evaluations for Manatee-Sarasota
Counties, Mid-Project Summary: Projects 1-11. Report to ManaSota 88, Pal-
metto, FL.
Dahl, I.E. and CE. Johnson. 1991. Status and Trends of Wetlands in the Con-
terminous United States, Mid-1970s to Mid-1980s. U.S. Department of the In-
terior, Fish and Wildlife Service, Washington, DC.
Devore, J. and R. Peck. 1986. Statistics: The Exploration and Analysis of
Data. West Publishing Company, St. Paul, MN.
Ehnenfield, J,G. 1983. The effects of changes in land use on swamps of the
New Jersey pine barrens. Biological Conservation 25:353-375.
Erwin, K.L. 1991. An Evaluation of Mitigation in the South Florida Water
Management District. Volume I. South Florida Water Management District.
West Palm Beach, FL
Erwin, K.L. 1990. Freshwater marsh creation and restoration in the Southeast,
p. 233-266. In J.A.-.Kuslerjand.M.£. Kentula (Eds.), Wetland Creation and
Restoration: The Status of the Science. Island Press, Washington, DC.
Erwin, K.L. 1988. A quantitative approach for assessing the character of fresh-
water marshes and swamps impacted by development in Florida, p. 295-297.
In J.A. Kusler, M.L. Quammen, G. Brooks (Eds.), Proceedings of the National
Wetland Symposium: Mitigation of Impacts and Losses. Association of State
Wetland Managers, Berne, NY. ',
Faulkner, S.P. and'Cj^Richardson. 1991. Physical and Chemical Characteris-
tics of Freshwater WetTand"Soils, p. 41-72. ,ln D.A. "Hammer (Ed.), Construct-
ed Wetlands for Wastewater Treatment. Lewis Publishers, Inc, Chelsea, Ml.
Federal'lnteragency.Commltte_e for Wetland Delineation. 1989. Federal Man-
ual for Identifying and Delineating Jurisdictional Wetlands. U.S. Army Corpj
of Engineers/ U.S. Environmental Protection Agency, U.S. Fish and Wildlife
Service, and U.S. Department of Agriculture Soil Conservation Service, Coop-
erative Technical Publication. Washington, DC.
-------
! .i "i. i'lijF IH
lllli: . ...... niii-s ...... ! !, '• 'K lint ..... ¥195" "ST"! ........ .'flBPirir ..... rsis»!Mt!W '
*; ..... ilSi'ii- SIMS 'i lint ..... i! ....... •oiii ti, T ••
„' ' • ,;,• i' • "f,, ii. '„!;•!" Ku, i ',, ' ,;,• nil „ ' '1 ", • " S: 'i .fi ' '! "i. ,r i
Gwin, S.E. and M.E. Kentula. 1990. Evaluating Design and Verifying Compli-
ance of Wetlands Created Under Section 404 of the Clean Water Act in Ore-
gon. EPA/6d6/3-90/061. U.S. Environmental Protection Agency, Environmen-
:al Research Laboratory, Corvallis, bR.
Gwin, S.E., M.E. Kentula, and D.L. Frostholm, in"conjunction"with R.L. Tighe.
1991. Evaluating;'Design and Verifying Compliance of Created Wetlands in
:he Vicinity of Tampa, Florida. EPA/600/3-91/068. U.S. Environmental Pro-
Action Agency, Environmental Research Laboratory, Corvallis, OR.
Hammer, D.A. 1992. Creating Freshwater Wetlands. Lewis Publishers, Inc.,
Chelsea, Ml.
138
-------
Hammer, D.A. (Ed.). 1989. Constructed Wetlands for Wastewater Treatment
Municipal, Industrial, and Agricultural Lewis Publishers, Inc., Chelsea, ML ..
Hitchcock, C.L and A. Cronquist. 1981. Flora of the Pacific Northwest—Ai
Illustrated Manual. 5th ed. University of Washington Press, Seattle, WA.;. .-
Holland, C.C and M.E. Kentula. In press. Impacts of.SectiOn 404 permits re
quiring compensatory mitigation on v/etlands in California. ;Wetlands Ecology
and Management. , r .-,•.'.•:;••.•'•• ;,-.••.• : '
Holland, C.C. and M.E. Kentula. 1991. The Permit Tracking System. (PTS): /
User's Manual. EPA/600/8-91/054. U.S. Environmental Protection Agency, En
vironmental Research laboratory, Corval I is, OR. ... ,'
Hollands, G.G. 1990: Regional analysis of the creation.and, restorationof-ket
tie and pothole wetlands, p. 281-298. In J.A. Kusler and M.E. Kentula (Eds.)
.Wetland Creation and Restoration: The Status of the Science. Island Press
.Washington, DC. • : •'•.... . ::•••• : ' : •• •• •'-• • ;
Hook, D.D., W.H. McKee Jr., H.K. Smith, J. Gregory, V.G. Burrell Jr., M.R
DeVoe, R.E. Sojka, S. Gilbert, R. Banks, LH. Stolzy,-C. Brooks, T.D. Matthew
and T.H. Shear. .1988. The Ecology and Management of .Wetlands. Timbe
Press, Portland, OR. . : ; . . - • ,^ .. : ;
Homer, R.R. and K.J. Raedeke. 1989. Guide for Wetland Mitigation Project
Monitoring. Report Number WA-RD 195.1. Washington State Department.q
Transportation, Seattle, WA. •• . • •'..-. ; - V ..
Hughes, H.G. and T.M. Bonnicksen (Eds.). 1990. Restoration '89: The Nev
Management Challenge. Proceedings of the First Annual Meeting of the :Soci
ety for Ecological Restoration, Oakland> CA. : . ;
Hughes, R.M., E. Rexstad, and C.E. Bond. 1987. The relationship of aquatii
ecoregions, river basins, and physiographic provinces to the ichthyogeograph
ic regions of Oregon. Copia 1887: 423-432. ;
Jordan, W.R., M.E. Gilpin, and J.D. Aber (Eds.). 1987. Restoration Ecology: /
Synthetic Approach to Ecological Research. Cambridge University Press
Cambridge, UK.
•.139
-------
_,_c, R.H. 1988. Monitoring wetland responses, p. 114-120. In J. Zelazny
"5 J.S. Feierabend (Eds.). Proceedings of Conference: Increasing Our Wet-
ancl Resources/National Wildlife Federation, Washington, DC.
i j
lantrud, H.A., Gl. Krapu, and G.A. Swanson: f989; Prairie Basin Wetlands
if the Dakotas: A Community Profile. U.S. Fish and Wildlife Service. Biolog-
cal Report 85(7.28). Washington, DC.
Celtula, M.E., J.C. Sifneos, J.W. Good, M. Rylko7 and K. Kunz! 1992. Trends
nd patterns in Section 404 permitting requiring compensatory mitigation in
Oregon and Washington. Environmental Management 16:109-119.
•'! i
Cing, blM. 1991 a. Wetland Creation and Restoration: An Integrated Frame-
vprk' for Evaluating Costs, Expected Results and Compensation Ratios. Pre-
|fed for U.S. Environmental Protection Agency; Office of Policy, Planning,
'.nlci Evaluation, Washington, DC.
Cin'g, D.M. 1991 b. Economics: Costing out restoration. Restoration & Man-
igfment Notes 9(1): 15-20.
Crgbs, Cj. 1989. Ecological Methodology. Harper and Row, New York, NY.
Cruczynski, W.L 1990. Options to'be considered in preparation and evalua-
ion of mitigation plans, p. 555-570. In J.A. Kusler and M.E. Kentula (Eds.),
Vetland Creation and Restoration: The Status of the Science. Island Press,
Yfshington, DC.
:::. Cusler, J.A. and M.E. Kentula (Eds.). 1990a. Wetland Creation and Restora-
ion: The Status of the Science. Island Press, Washington, DC.
- • • ' ;• j - !
Cusler, J.A. and M.E. Kentula. 1990b. Executive Summary, p. xvii-xxv. In J.A.
Cusler and M^E. Kentula (Eds.), Wetland Creation and Restoration: The Status
>f the Science. Island Press, Washington, DC.
Cusleo J.A. and S. Daly (Eds.). 1989. Proceedings of an International Sympo-
[ujn: Wetlands and River Corridor Management. Association of State Wet-
and Managers, Berne, NY.
.angis, R., M. Zalejko, and J.B. Zedler. 1991. Nitrogen assessments in a con-
tructed and a natural salt marsh of San Diego Bay, California. Ecological Ap-
»lications 1:40-51.
140
-------
Larsen, D.P., J.M. Omernik, R.M. Hughes, C.M. Rohm, T.R. Whittier, A.J. Kin-
ney, A.L. Gallant, and D.R. Dudley. 1986. The correspondence between spa-
tial patterns in fish assemblages in Ohio streams and aquatic ecoregions. Envi-
ronmental Management 10:815-828.
Leibowitz, N.C., L Squires, and J.P. Baker. 1991. Research Plan for Monitor-
ing Wetland Ecosystems. EPA/600/3-01/010. U.S. Environmental Protection
Agency, Environmental Research Laboratory, Corvallis,
Leibowitz, S.G., E.M. Preston, L.Y. Arnaut, N.E. Detenbeck, C.A. Hagley, M.E.
Kentula, R.K. Olson, W.D. Sanvilte, and R.R. Sumner. 1991 Wetlands Re-
search Plan FY92-96: An Integrated Risk-Based Approach. EPA/600/R-
92/060. U.S. Environmental Protection Agency, Environmental Research Lab-
oratory, Corvallis, OR.
Leopold, Aldo. 1966. Sand County Almanac. Oxford University Press. New
York, NY.
Ludwig, J.A. and J.F. Reynolds. 1988. StatisticalEcoIogy: A Primer on Meth-
ods and Computing. John Wiley and Sons, New York, NY.
Lugo, A.E., S. Brown, and M. Brinson. 1990. Forested Wetlands. Ecosystems
of the World 15. Elsevier, Amsterdam, The Netherlands.
Lyons, J. 1989. Correspondence between the..distribution offish assemblages
in Wisconsin streams and Omernik's ecoregions. American Midland Natural-
ist 122:163-182.
Marble, A.D. 1990. A Guide to Wetland Functional Design. Report Number
FHWA-IP-90-010. Federal Highway Administration, McLean, VA.
Marsh, W.M. 1978. Environmental Analysis for Land Use and Site Planning.
McGraw-Hill Book Company, New York, NY.
Majumdar, S.K., R.P. Brooks, F.J. Brenner, and R.W. Tiner, jr. (Eds.). 1989.
Wetlands Ecology and Conservation: Emphasis in Pennsylvania. The Pennsyl-
vania Academy of Science, Philadelphia, PA. .
McVoy G.R. 1988. Advantages of open water/emergent wetlands for mitiga-
tion .and a holistic approach to banking, p. 289-290. In J.A. Kusler, M.L.
Quammen, and G. Brooks (Eds.), Proceedings of the National Wetland Sympo-
141
-------
sium: Mitigation of Impacts and Losses, Association of State Wetland Man-
agers, Berne, NY.
! is I | .';', .• ; ' ••! ' ; •.;..,'. " • :'!.> • ill .. , it; j.:,' ' '•. '^ ', • : : :>i iyj;ff «V:
MItsch, WJ. and J.G. Gosselink. 1986. Wetlands. Van Nostrand Reinhold
Company, Inc., New York, NY.
Mjtsch, WJ., M. Straskraba, and S.E. Jorgensen (Eds.). 1988. Wetland Model-
ling. Elsevier Science Publishing Company, Inc., New York, NY.
.
Murkin, H.R. (Ed.). 1984. Marsh Ecology Research Program Long-term Moni-
toring Procedures Manual. Delta Waterfowl Research Station, Manitoba,
Canada.
National Research Council, Committee on Res
oration of Aquatic
Ecosystems—Science, Technology, and Public Policy, Water Science and
technology Board, Commission on Geosciences, Environment, and Resources.
1992. Restoration of Aquatic Ecosystems: Science, Technology, and Public
Policy. National Academy Press, Washington, DC
Neter, J., W. Wasserman, and M.J. Kutner. 1990. Applied Linear Statistical
Models: Regression, Analysis of Variance, and Experimental Designs. IRWIN,
Homewood, IL.
Newling, CJ. and Landin, M.C. 1985. Long-term Monitoring of Habitat De-
velopment at Upland and Wetland Dredged Material Disposal Sites, 1974-
1982. Technical Report D-85-5. Environmental Laboratory. U.S. Army Engi-
neers Waterways Experiment Station, Vicksburg, MS.
Niering, W.A. 1991. Wetlands of North America. Thomasson-Grant, Char-
lottesville, VA.
i! S, I {(IH , . . •.:•:• ' • •, .'..,..•!•..• :!':
-------
Alteration in the Glaciated Northeastern, United States: An Assessment of .the
Science Base. Publication 87-1. Environmental Institute, University of Massa-
chusetts, Amherst, MA. j
Omernik, J.M. 1987. Ecoregions of the conterminous United..States. Annals
of the Association of American Geographers 77(1 ):118-125. (map scale
1:7,500,000) . , . '. , •;'••.,. •
Owen, C.R. 1990. Effectiveness of Compensatory Wetland Mitigation in Wis-
consin. Technical Report to the Wisconsin Wetlands Association, The Lake
Michigan Federation, The American Glean Water Project. .University of Wis-
consin, Madison, Wl. , .
Qwen/CR., Q.J. Carpenter, and C.B. DeWitt. 1989. - Evaluation;.of Three
Wetland Restorations Associated with. Highway Projects. Technical Report,
Institute for Environmental Studies, University of Wisconsin, Madison/ Wl. v/;
Pacific Estuarine Research Laboratory. 1990. A .Manual for Assessing Re-
stored and Natural Coastal Wetlands with Examples from Southern California.
California Sea Grant Report Number T-CSGCP-021. Lajolla, CA. r >:•
Pielou, E.C 1986. Assessing the diversity and composition of restored vegeta-
tion. Canadian Journal of Botany 64:1344-1348.
Quammen, M.L. 1986., Measuring the sujccess of wetlands: mitigation. Na-
tional Wetlands Newsletter 8(5):6-8. -, ;
Reed, P.B., Jr. 1988. National List of Plant Species That Occur in Wetlands:
Northwest (Region 9). U.S. Fish>and Wildlife Service. Biological .Report
88(26.9). Washington, DC. (This cite refers to the Pacific Northwest Volume.-
Other volumes are appropriate for specific areas).
Roberts, T.H. 1991. Habitat Value of Man-made .Coastal Marshes, in Florida.
Technical Report WRP-RE-2. U.S. Army Engineers Waterways Experiment Sta-
tion, Vicksburg, MS. . '.•:•"
Roethke, Theodore. 1961. Words..for the. Wind. Indiana University Press,
Bloqmington, IN. : ; ; ' :' . •-•
Rohm, CM., J.W. Giese, and C.C. Bennett. 1987. Evaluation of an aquatic
ecoregion classification of streams in Arkansas. Freshwater^ Ecology 4:127-
140.. • . - . - .-;..;• --. , \.v .-;.-•'. •.•-:...: - - '• v •
143
-------
It''"
ii'Li
Schemnitz, S. 1980. Wildlife Management Techniques Manual. 4th ed. The
Wildlife Society, Washington, DC.
Schneller-McDonald, K., LS. Ischinger, and G.T. Auble. 1989. Wetland Cre-
ation and Restoration: Description and Summary of the Literature. U.S. Fish
arid Wildlife Service, BiologicarReporf: 89. Washington, DC.
Segeiquist, cX W.L. Slauson, M.L. Scott, and G.T. Auble. 1990. Synthesis
of Soil-Plant Correspondence Data From Twelve Wetland Studies Throughout
trie United States. U.S. Fish and Wildlilfe Service, Biological Report 90(19).
Washington, DC.
i ' | ...... • , ..... - .' ,, ;, ,,, . Jl'V .;*;"'!,•;;»; :Ji>ilU' ..... t liv' ..... '"'•' "1 *>•;;'•
-------
the National Wetland Symposium: Wetland Hydrology. Association of State
Wetland Managers, Berne, NY.
Snedecor, G.W. and W.G. Cochran. 1980. Statistical Methods. The Iowa
State University Press, Ames, I A.
Sokal, R.R. and F.J. Rohlf. 1981. Biometry. W.H. Freedman and Company,
New York, NY.
Stark, N. 1972. Low Maintenance Vegetation: Wildland Shrubs, Their Biolo-
gy and Utilization. General Technical Report INT-1. U.S. Department of Agri-
culture, Forest Service. Washington, DC. .
Steward, A.N., LJ. Dennis, and H.M. Gilkey. 1963. Aquatic Plants of the Pa-
cific Northwest. 2nd ed. Oregon State University Press, Coryaliis, OR.
Thornburg, A. 1977. Use of vegetation for stabilization of shorelines of the
Great Lakes, p. 39-53. In Proceedings of the Workshop on the Role of Vegeta-
tion in Stabilization of the Great Lakes, Shoreline. Great Lakes Basin Commis-
sion, Ann Arbor, Ml. . '
Tiner, R.W., Jr. 1984. Wetlands;"6f the United" Stales: Current Status arid Re-
cent Trends. U.S. Fish and Wildlife Service, National Wetland Inventory,
Washington, DC.
Tiner, R.W., Jr. 1988. Field Guide to Nontidal Wetland Identification. Mary-
land Department of Natural Resources, Annapolis, MD arid U.S. Fish and
Wildlife Service, Newton Corner, MA. Cooperative publication.
U.S. Department of Agriculture, Soil Conservation Service. 1992. Field Hand-
book. Chapter 13: Wetland Restoration, Enhancement, and Creation. Wash-
ington, DC.
U.S. Fish and Wildlife Serviced 1980. Habitat Evaluation Procedures. ESM
102. U.S. Department of the Interior, Fish and Wildlife Service, Division of
Ecological Services, Washington, DC. -\
van der Valk, A.G. (Ed.). 1989. Northern Prairie Wetlands. Iowa State Uni-
versity Press, Ames, I A.
van der Valk, A.G. and C.B. Davis. 1976. The seed banks of prairie glacial
marshes. Canadian journal of Botany 54:1832-1838.
145
-------
'*' Uillii ''''i I !!! II
Jl'" IE i! .I1: I
IIP" 'ill. I,"1:, i 11
iiiiiii" iiii1'!1' • • i1. i
':li,r ,lt IP 'I ' ll<
Hfl1 iiif1 i ' i
Verner, S.S. 1990. Handbook for Preparing Quality Assurance Project Plans
for Environmental Measurements. Technical Resources, Inc., Rockville, MD.
! ! ' , ,:, <:i"i!;; !i 'lilll-H ' ,:••;', i . 'fi '. >i ' ' ' ."' "' .:'"'' ' ,.'•': "''f'i '.: • i I II ' ''"•.', " "]„'::
,7'Weinhpld, C.E. and A.G. van der Valk. 1988. The impact of duration of
Hrainage On the seed banks of northern prairie wetlands. Canadian Journal
Botany 67:1878-1884.
Weller, M.w;, G:W. Kaufmann; and P.A. Vohs. 1991. Evaluation of wetland
development and waterbird response at Elk Creek Wildlife Management Area,
Lt .t^ / '.'.'..
yyentwdrth, t-R., G.P. Johnson, and R.L. Kologiski. 1988. Designation of wet-
||nds by weighted averages of vegetation data: a preliminary evaluation.
Water Resources Bulletin 24(2):389-396.
White, T.A., R. Lea, R.J. Haynes, W.L. Nutter,. J.R. Nawfot, M.M. Brinson, and
A.F. Clewell. 1990. Development and summary of MiST: a classification sys-
tem for preproject mitigation sites and criteria for determining successful repli-
cation of forested Wetlands, p. 323-335. In J. Skousen and J. Sencindiver
(Eds.), Proceedings of the 1990 Mining arid Reclamation Conference and Exhi-
bition. American Society of Surface Mining and Reclamation, Charleston, WV.
Whittier, T;R.; R.M. Hughes, and D.P. Larseri: 1988. Correspondence be-
tween ecoregions and spatial patterns in stream ecosystems in Oregon. Cana-
dian Journal of Fisheries and Aquatic Science 45:1254-78.
Whittier, T".R., D.P. Larsen, R.M. Hughes, C.MTRohrn,A.L. Gallant/and J.M.
Omernik. 1987. Ohio Stream Regionalization Project: A Compendium of Re-
sults. Freshwater Research Laboratory, Corvallis, OR.
':.'•!Winter/T.C.' 1981. Uncertainties in estimating the water balance of lakes:
Water Resources Bulletin. 17(1): 82-115.
ledleo j.B. and M^Kentula. 1986. Wetlands Research Plan. EPA/600/3-
86/009:- u;s. Envirorimehtal Protection Agency, Environmental Research Lab-
oratory, Corvallis, OR. - :
' •' ' ' ' 's f. ' '" ij j
ZedlerJ.B. and R. Lahgis. 1991. 'Comparisons of constructed and natural salt
marshes of Sari Diego Bay. Restoration and Management Notes 9(1 ):21 -25.
146
-------
INDEX
1990 Farm Bill, 29
Alabama, 12,144
areas at risk, 13,24
Arkansas, 5,12,17,143,144
as-built assessment, 44, 52, 56, 57,
61
as-built conditions, 44, 52, 59-61,
64
assessment, i, 12,17,43,44,52,56,
57, 59-61, 66-69, 72,143, 144
procedures, 23, 37,43,44,59,
60, 66> 68-70, 72, 80, 81, 96,
142,145
base map, 56, 61
basin, 111,113,114,116-118,130,
140,145
buffers, 116,132
California, 12,17, 26,138-140,143
characterization curves, 87, 93> 96,
107 "~~"
Clean Water Act, 11, 74, 138
Section 401,15
Section 404, xi, 12,13,15,17,
19,24,35,74,112,138-140,
144
comparability, 69, 70, 80
compensatory mitigation, 1,13,17,
26,139,140
compiling information, 13,15,29
compliance, 2,17, 43, 44, 52, 72,
130,138
comprehensive assessment, 17, 57,
60
confidence interval, 97,102
Connecticut, i, 88,105,112,117,
127,131,136,144
construction plans, 44, 52,127,130
contouring, 114,117 .
criteria, 2, 3, 8, 9, 36,37, 44, 57,
59, 60, 63, 69, 72, 87, 98,102,
104,105,108,131,146
Dakotas, 130,140
data, i, xi, 3,11 -13,15,17,19,24,
26,29,43,44, 52, 56,57, 59-61,
64, 66, 68-76, 78-82, 87, 88, 92,
93, 96-98,100,102,103-105,
107,108,112,113,129-131,
135,137,144,146
analysis, 3,11,13,15,26, 60, 68,
96,98,112,120,137,139,
.141,142
collection/44, 52, 56, 57, 59, 60,
68, 69, 71, 74-76, 78-82, 87, 98
entry, 15,68,98
management system, 13,15
retrieval, 13
design, 3, 5, 8, 9, 24, 44, 52, 57, 59,
.60, 70-72, 75, 77, 97,111,113,
114, 117,127,129-131,138, 141
guidelines, 3,, 9, 72,111,120,
123,135
dominance measure, 124
dominant species, 64,123,124
donor wetland, 122,123, 130
ecological setting, 5, 33, 35, 37, 75,
132
ecoregion(s), 5,17, 33, 35,139,
141,143,146
emergent marsh, 29; 105,112,120,
125,131
endemic, 125
erosiorVri, 122"-124
evaluation, 2, 7,12,143,44, 52, 59,
60, 63, 66, 68, 74, 76, 87, 88, 92,
105, 131,137,140, 143,145, 146
exotics, 119,125
fauna, 61,66, 77,107
field study ;
Connecticut Study, 105,127
-Florida Study, 29, 35, 92,120
147
-------
I1:1.,
Ifffif'i ''1, J;,,
t«•;•!,i (•••:•..
"It PI" •' i i, I 1
li'jh IllliW "!i I
IH'M I ! I!
"
Oregon Study, 29, 33, 35-37, 52,
92,98,103,113,114,119,
120,130
flood detention, 112
Florida, i, 26, 29, 35, 88," 92, 1 05,
107,113,117,118,120,131,
136,137,138,143
freshwater nontidal wetlands, 3
grOUhdwater, 117, 129
growing season, 66, 69, 71 , 1 27
habitat, 1 9, 52, 56, 66, 71 , 73, 97,
112,113,137,142-145
Habitat Evaluation Procedures
(HEP), 66
herbaceous, 64, 1 00, 1 02, 1 03, 1 22,
124,125
homogenous, 29, 33, 97, 98, 1 00
hydrology, xi, 8; 44, 52, 61 , 63, 71 ,
76,103,113,118,120,122,127,
,129,142,144
hydroperiod, 118,127,129
hydrophytes, 8, 1 1 7
Impacted wetlands, 1 2, 1 3, 1 5, 1 1 1 ,
::;ii2 ' .". ; • ........ .-. • : .......... ":"
indicator(s), 7-8, 23, 43, 57, 63, 66,
92,93,103,135
inland wetlands, 1 35
invasive, 119, 125
jurisdictional wetland, 63, 93
land use, 5, 9, 3 5, 3 7, 52, 61 , 88,
135 137 141
landscape, i;l 35, 60, 67, 68, 112,
'' ' ..... " "
...... , .
long-term research, i, 59
Ipuisiana, 12, 1 7, 26, 144
maintenance, 1 23, 1 45
Capping, 56, 61,74, 76
maps, 33, 36, 37, 52, 56, 61 ,112
metals, 67, 131
Mississippi, 12, 144
mitigation, i, xi, 1 -3, 7, 1 2, 1 3, 1 5,
17, 23, 24, 26, 29, 33, 35, 72, 74,
: 75, 79,113,131,135,137, 139-
143,146
moisture gradient, 118
monitoring, 2, 3, 7, 8,13,17, 23,
24, 26, 37, 43, 44, 52, :59-61, 63,
66, 67, 71-75, 79, 82, 87, 92, 96,
97,108,123,135,139-142
morphometry, 44, 61,63
mulch, 100,104,122,123,131
National Wetlands Inventory, 26
nitrogen, 130,140
nitrogen fixation, 130
nurseries, 124,127
nutrients, 64, 67,130
open water, 29, 36, 52, 92,112,
113,127,129,141
opportunistic species, 125
Oregon, i, xi, 5,12,15,17, 26, 29,
33, 35-37, 52, 73-75, 88, 92, 98,
103,105,107,111-114,117-120,
127, 129-131, 138-140, 145, 146
organic, 63, 64, 88, 93, 96, 97,100,
107,122,130,131
matter, 63, 64, 88, 93, 96, 97,
100,107; 130,131
soils, 5, 8, 56, 63, 70,76,103,
120,122,127,130,131,135,
137
outliers, 92,100,104
paired wetlands, 105
palustrine, 17,36,112,117,120,
125,136
emergent marshes, 36,120
emergent wetland, 17,119,120
forested wetland, 17
percent, 63; 64, 88, 92, 93, 96-98,
100,102,104,107,124,130
of open water, 92,112,113,141
of species in common, 102 , .
organic matter, 63, 64, 88, 93, 96,
97,100,107,130,131
performance, 2,3, 5, 7-9,23, 24,
"lililir iiii'iil '::
'I IB ',
A Pill5
148
-------
35, 37, 43, 44, 52, 56, 57, 59, 60,
69, 72, 87, 88, 92, 93, 96-98,
100,102-105,108,131,132
curves, 5, 7-9, 23,24, 87, 88, 93,
96,104,105,107,131
permanent sampling plots, 68
permit, xi, 2, 3,12,13,15,17,19,
26, 37, 43, 44, 52, 60, 61, 66, 68,
72,112,120,139
conditions, 5,33, 35, 37,43, 44,
52,56,59-61,63,64,66-72,
74,120,122,124,129
record, 12,13, 15, 26, 52, 56, 57,
61, 63, 67, 68, 76, 80,112,114
specifications, 17,127
tracking system, xi, 13,139
permitting, 11 -13,15,17,19, 24,
29,44,56,59,140,144
activity, 11,13,17, 24, 26, 64, 72
agencies, 1,11,29, 73, 74, 77,
81,87
assessment of the effects of, 12
cumulative impacts of, 11,13
systems, 11,15,35, 61, 82, 92,
113,130,132
trends in, 11-13,19,26, 108 " "~~
plant community, 64, 66, 124
composition, 63, 66,104,105,
118,120,123,143
cover, 33, 64, 98,100,102-104,
120,124,135
planting lists, 118, 119
pond(s), 98, 105,112-114,116,
118,129,131
Portland, Oregon, xi, 17, 26, 29733,
35, 36, 98, 103, 104,112,138,
139
post-construction monitoring, 43
precision, 29, 36, 69, 71
progressive mean, 36
quality assurance objectives, 60
red maple swamps, 130, 138
reference sites, 3,5, 35
region, 11, 35, 57, 71, 72, 74, 75,
112-114,120,124,125,138,
143, 144
regional flora, 125
regression, 96, 97; 142
regulation, 1,11,138
regulatory decisions, i, 1,17
relative, 5,9,33, 35,114,120
abundance, 24, 63, 66, 68,120,12^
elevations, 113,114
representative sample, 26, 36
representativeness, 69, 70
restoration, 2, i, ii, xi, 2,3, 5, 7, 9,
24, 35, 60, 69, 71, 92,111, 114,
108,120,123,127,131,132, ;
135, 137-140, 142, 144-146
revegetation, 120,122
Rhode Island, 130
riparian system, 120
risk, 11, 13, 24, 26, 29, 33, 37, 92,
141
Rjvers and Harbors Act, Section 10,
11
routine assessment, 57, 59, 60
sample, 23,24,26, 29, 35, 36, 59,
66, 70, 71, 88, 92, 97, 98,100,
102,104,105,114
sample size, 29, 70
stratified, 35, 70
sampling, 23, 24, 26, 36, 37, 60, 64,
66, 68-72, 80, 88, 100, 104,123
design, 70
- efficiency, 68, 69
protocols, 60, 66
strategy, 3,17, 23, 37, 68-70, 72,
123
saturated, 63,129
Seaside, Oregon, 75
sediment retention, 57, 112
seed bank, 123
seeds, 122,124,131 .
149
-------
setting priorities, 23
Shrub/scrub wetland, 120
site selection, i, 5, 29,37, 70
slope(s), 61,63, 96198,112-114,
116-118,120
Society for Ecological Restoration,
139 , ,
soil, 29, 63, 64, 71,$3, 97,107,
116,117,122,125,130,131,
137/144,145
augmentation, 122,131
gleyed, 63
hydric, 8, 56, 63, 71
microflora", 122
mineral, 131
mottles, 63
organic matter, 63, 64, 88, 93, 96,
97,100,107,130,131
pore water, 122
saturation, 63
.stabilization, 125,145
species, xi, 7,19, 52, 60, 64, 66, 68,
71, 73, 88, 98, 100,182-105,
118,119,120,122-125,127,
131,143
composition, 63, 66,104,105,
118,120,123,143
diversity, 7, 8,43, 88,100,104,
105,143
diversity index, 88
plant diversity, 100,105
sprigs, 124
staff gauge, 129
standard operating procedures, 60
state-wide standardization, 13
statistical tests, 96,108
~ Rest/97
hypothesis tests, 97
Levine's test, 97
statistical analyses, 96
statistics, 137 - '. ;
Student's T-test, 96, 98
I II I
structural characteristics, 112,116,
127
study area, 26, 33, 36
substrate(s), 44, 52, 59, 61', 63, 64,
88, 96, 97, 100, 122,125,127,
129-131 .
success, 2, 3, 71-74, 76-80, 87,108,
123; 127,132,143
successional species, 125
surface water, 63,105
Tampa, Florida, 138
Texas, 12,17,19/144 '
timing of sampling, 26
topographical profiles, 114
training, 56, 63, 69, 70, 76, 79-81
transects, 61, 70, 71, 98,114
transitional area, 113,116-118
trends in permitting, 11,13,19
upland, 71,103,114,116-118,142
variability, 3,29, 52, 57, 67, 70, 92,
97,100,118
variance, 142
variable, 8, 63, 69, 70, 88, 93,105
vegetation, 5> 44, 52, 56, 61, 64; 66,
69-71, 76, 98, 100, 102-104,107,
111,112-114, 118-120, 122-125,
127,130,143,145,146
communities, 56, 59, 63, 64, 66,
69,75,98,118-120,123,135
cover, 98,103
e'mefgerit vegetation, 98,118
herbaceous vegetation, 100, 103
percent cover, 98,100,104,124
stratum, 70,124
zonation, 118
volunteer species, 119,120
Washington, 12,17, 73,124,135-
140,142-145
water, i, 11,29, 36, 43, 52, 57, 61,
63, 67, 68, 70, 71, 74, 76, 80, 92,
'-'••• -'93797/fte, 107, 111-114,116,
117,122,125,127,129,130,
150
-------
135-138,141-143,146
budgets, 129
control structures, 52
depth, 59, 63, 68,113,117,127,
130
level, i, 1,2,5, 7-9,36,44,60,
63, 68, 69, 75, 79, 82, 88, 92,
93, 97, 98, 105,107,116,122,
129,130,135
level recorders, 129
quality, 5, 7,12,13, 35, 36, 60,
67, 69, 72, 73, 75, 76, 79, 80,
82,97,130,135,144,145
retention, 57,112,130,131
source, 69, 96,116,117,122,
130
table, xi, 12,13,15, 29, 36,44,
61,63,68,107,113,116,117,
125,127,132
waterfowl, 29,113,142
watershed, 5, 61, 93
weedy species, 131
weighted average, 92,103
wetland management, 1, 7, 11, 24,
71,108
management decisions, 1,3,11,
12,15,17,19,82
managers, i, 3, 8,13,17, 59, 74,
82, 87, 92, 96, 108, 135-137,
140,142,144
Willamette Valley, Oregon, xi, 112
Wisconsin, 5,111,113,141 -143
woody cuttings, 124
woody species, 122,125
»U.S. GOVERNMENT PRINTING OFFICE: l992-6t8-oo»60039
151
-------
j
'I ifl •.
'J, I1 Slift '
fit!,
------- |