EPA-843-B-24-003
National Wetland Condition Assessment 2021
Technical Support Document
US Environmental Protection Agency
Office of Water
Office of Research and Development
Washington, DC
-------
This page was intentionally left blank for double-sided printing.
-------
Notice
The National Wetland Condition Assessment 2021: Technical Support Document (EPA 843-B-24-003)
details methods and analysis approaches used in the 2021 National Wetland Condition Assessment
(NWCA) conducted by the United States Environmental Protection Agency (USEPA) and partner
organizations. This document supports the NWCA results presented in National Wetland Condition
Assessment: The Third Collaborative Survey of Wetlands in the United States (EPA-843-R-24-001).
The information in the Technical Support Document has been funded wholly or in part by the US
Environmental Protection Agency. This technical report has been subjected to review by the USEPA Office
of Water and approved for publication. Approval does not signify that the contents reflect the views of
the Agency, nor does mention of trade names or commercial products constitute endorsement or
recommendation for use.
The suggested citation for this document is:
US Environmental Protection Agency. 2024. National Wetland Condition Assessment 2021:
Technical Support Document. EPA 843-B-24-003. US Environmental Protection Agency, Washington,
DC.
Companion documents for the NWCA are:
National Wetland Condition Assessment 2021: Quality Assurance Project Plan (EPA-843-B-21-004)
National Wetland Condition Assessment 2021: Site Evaluation Guidelines (EPA-843-B-21-001)
National Wetland Condition Assessment 2021: Field Operations Manual (EPA-843-B-21-002)
National Wetland Condition Assessment 2021: Laboratory Operations Manual (EPA-843-B-21-003)
National Wetland Condition Assessment: The Third Collaborative Survey of Wetlands in the United
States (EPA-843-R-24-001)
If you decide to print the document, please use double-side printing to minimize ecological impact.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
1
-------
This page was intentionally left blank for double-sided printing.
National Wetland Condition Assessment: 2021 Technical Support Document
Dec 2024
-------
Acknowledgements
The EPA Office of Water (OW) and Office of Research and Development (ORD) would like to thank the
many people who contributed to the National Wetland Condition Assessment 2021. Without the
collaborative efforts and support by state and tribal environmental agencies, federal agencies,
universities, and other organizations, this groundbreaking assessment of wetlands would not have been
possible. In addition, the survey could not have been done without the innumerable field biologists, soil
scientists, taxonomists, statisticians, and data analysts, as well as program administrators, EPA regional
coordinators, project managers, quality control officers, and reviewers. To the many hundreds of
participants, EPA expresses its profound thanks and gratitude.
State and Tribal Agency Partners
Alabama Department of Environmental Management
Arizona Department of Environmental Quality
California State Water Resources Control Board
Colorado Natural Heritage Program
Confederation of Northern Mariana Islands Bureau of
Environmental and Coastal Quality
Confederated Tribes of the Colville Reservation
Confederated Tribes of the Umatilla Indian Reservation
Delaware Department of Natural Resources and
Environmental Control
District of Columbia Department of Energy and
Environment
Florida Department of Environmental Protection
Georgia Department of Natural Resources
Guam Environmental Protection Agency
Idaho Department of Environmental Quality
Illinois Environmental Protection Agency
Illinois Natural History Survey
Indiana Department of Environmental Management
Iowa Department of Natural Resources
Kansas Department of Health and the Environment
Kansas Water Office
Kentucky Division of Water
Leech Lake Band of Ojibwe, Division of Resource
Management
Louisiana Department of Wildlife and Fisheries
Maine Department of Environmental Protection
Maine Natural Areas Program
Maryland Department of the Environment
Massachusetts Department of Environmental Protection
Michigan Department of Environment, Great Lakes, and
Energy
Minnesota Pollution Control Agency
Missouri Department of Natural Resources
Montana Natural Heritage Program
Navajo Environmental Protection Agency
Nebraska Game and Parks Commission
Nevada Division of Environmental Protection
New Hampshire Department of Environmental Services
New Jersey Department of Environmental Protection
New Mexico Environmental Department
New Mexico Natural Heritage Program
New York Natural Heritage Program
North Carolina Department of Environment and Natural
Resources
North Dakota Department of Health
Ohio Environmental Protection Agency
Oklahoma Conservation Commission
Oregon Department of Environmental Quality
Oregon Division of State Lands
Pennsylvania Department of Environmental Protection
Quinault Indian Nation
South Carolina Department of Health and Environment
Control
Tennessee Department of Conservation and Environment
Texas Commission on Environmental Quality
Utah Department of Environmental Quality
Utah Geological Survey
Vermont Department of Environmental Conservation
Virginia Department of Environmental Quality
Washington Department of Ecology
Washington Natural Heritage Program
West Virginia Department of Environment Protection
Wisconsin Department of Natural Resources
Wyoming Department of Environmental Quality
Wyoming Natural Diversity Database
Federal Partners
National Park Service
U.S. Army Corps of Engineers
U.S. Department of Agriculture, Natural Resource
Conservation Service
U.S. Fish and Wildlife Service
U.S. Forest Service
U.S. Geological Survey
U.S. EPA Office of Research and Development
U.S. EPA Office of Water
U.S. EPA Regions 1-10
3
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Additional Collaborators
Avanti
Burke Museum Herbarium
Coastal Environment
Crow Insight
Eastern Kentucky University
EnviroScience
ESS Group
Four Peaks Environmental Science and Data Solutions
General Dynamics Information Technology
Great Lakes Environmental Center
Midwest Biodiversity Institute
Moss Landing Marine Laboratories
New England Interstate Water Pollution Control
Commission
North Dakota State University
Oregon State University
PG Environmental
Riparia at Pennsylvania State University
Southern California Coastal Water Research Project
University of Central Missouri
University of Florida
University of Houston - Clear Lake
University of Illinois
University of Montana
University of New Mexico
University of Wyoming
Virginia Institute of Marine Sciences
Members of the Data Management and Analysis Team for the NWCA 2021 were Amanda Nahlik, Karen
Blocksom, Michael Dumelle, Tony Olsen, Teresa Magee (retired), Mary Kentula (retired), Anett Trebitz,
Tom Kincaid, and Marc Weber from USEPA Office of Research and Development; Gregg Serenbetz, Sarah
Lehmann, Danielle Grunzke, Garrett Stillings and Kerry Kuntz from USEPA Office of Water; and Alan
Herlihy from Oregon State University.
Key assistance in acquisition of plant species trait information or in taxonomic standardization of plant
species names was provided by Macy Carr, Tara Mazurczyk and Morgan Mack (Oak Ridge Institute for
Science and Education fellows at USEPA Office of Water).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table of Contents
Notice 1
Acknowledgements 3
Table of Contents 5
List of Figures 10
List of Tables 14
Acronym List 19
Chapter 1: Project Overview and Elements of Analysis 21
1.1 Overview 21
1.2 Objectives of the NWCA 21
1.3 Elements of Analysis 22
Chapter 2: Survey Design 25
2.1 Description of the NWCA Wetland Type Population 25
2.2 Sample Frame, Survey Design, and Site Selection 25
2.2.1 Sampling frame 25
2.2.2 Survey design units 26
2.2.3 Expected Sample Size 27
2.2.4 Survey design 28
2.2.4.1 Re-sampled Site Design 28
2.2.4.2 New Site Design 28
2.2.5 Number of Sites Expected to be Sampled 29
2.2.6 State-Requested Modifications to the Survey Design 30
2.3 Wetland Area in the NWCA Sample Frame 32
2.4 Survey Analysis 33
2.5 Estimated Wetland Extent of the NWCA Wetland Population and Implications for Reporting 33
2.6 Literature Cited 34
Chapter 3: Selection of Handpicked Sites 37
3.1 Pre-Sampling Selection of Handpicked Sites 37
3.1.1 Initial Screen 37
3.1.2 Basic Screen 37
3.1.3 Landscape Screen 39
3.1.4 Distribution of Handpicked Sites 39
3.1.5 Replacement of Handpicked Sites Not Sampleable 39
3.1.6 Results 40
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
3.2 Literature Cited 41
Chapter 4: Data Preparation 43
4.1 Data Entry and Review 43
4.1.1 Field Data 43
4.1.1.1 Field Data Validation 43
4.1.2 Laboratory Data 43
4.2 Quality Assurance Checks 43
4.2.1 Verification of Points 44
4.2.2 Confirmation of Coordinates Associated with the Sites Sampled 44
4.2.3 Data Checks 45
4.3 Literature Cited 45
Chapter 5: Subpopulations 46
5.1 Literature Cited 46
Chapter 6: Assigning Disturbance Class 53
6.1 Sites Used to Establish the Disturbance Gradient 54
6.2 Establishing a Disturbance Gradient 55
6.2.1 Indices and Metrics 55
6.2.2 Setting Least-Disturbed Thresholds 56
6.2.3 Setting Most-Disturbed Thresholds 56
6.2.4 Classifying Disturbance at Each Site for each Sampling Visit 56
6.3 Human-Mediated Physical Alteration Screens and Thresholds 57
6.4 Chemical Screens and Thresholds 58
6.5 Abiotic Disturbance Class Assignments 60
6.6 Biological Screen and Threshold 61
6.7 Final Disturbance Class Assignments 62
6.8 Literature Cited 64
6.9 Appendix A: Illustrative Guide to Assigning Disturbance Class in Six Steps 66
Chapter 7: Vegetation Analysis Overview, Data Acquisition, and Preparation 73
7.1 Background 73
7.2 Overview of Vegetation Analysis Process 74
7.3 Vegetation Data Collection 76
7.3.1 Field Sampling 76
7.3.2 Identification of Unknown Plant Species 79
7.4 Data Preparation - Parameter Names, Legal Values, and Data Validation 79
7.4.1 Description of Vegetation Field Data Tables 79
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.4.2 Data Validation 80
7.5 Nomenclatural Standardization 82
7.5.1 Nomenclature Reconciliation Methods 82
7.5.2 Nomenclature Standardization Results and Documentation 85
7.6 Species Traits - Life History: Growth-habit, Duration, and Plant Category 86
7.6.1 Growth-Habit 86
7.6.2 Duration 86
7.6.3 Plant Categories 87
7.7 Species Traits - Wetland Indicator Status 88
7.7.1 Wetland Indicator Status Assignment Process 89
7.8 Species Traits - Native Status 91
7.9 Species Traits - Coefficients of Conservatism 93
7.9.1 Compilation of Existing State and Regional C-Value Lists from Across the Conterminous US... 94
7.9.2 Assigning Existing C-values to Taxon-Region Pairs Observed in the NWCA Surveys 97
7.9.3 Defining C-values for NWCA Taxon-Region Pairs Where None Were Available 97
7.9.4 Final NWCA C-value Trait Table 98
7.10 Literature Cited 99
7.11 Appendix B: Parameter Names for Field Collected Vegetation Data 103
7.12 Appendix D: Existing Coefficient of Conservatism Lists included in the Compiled C-value Lists
{unpublished draft) assembled by NWCA 107
Chapter 8: Vegetation Analyses and Candidate Metric Evaluation Prerequisite to Multimetric Index
Development 115
8.1 Overview 115
8.2 Anthropogenic Disturbance 116
8.3 Considering Regional and Wetland Type Differences 117
8.4 Calculating Candidate Metrics 122
8.5 Evaluating Candidate Vegetation Metrics 123
8.5.1 Range Tests 124
8.5.2 Repeatability (S:N) 124
8.5.3 Responsiveness 125
8.5.4 Redundancy 125
8.5.5 Application of Metric Screening Criteria 126
8.6 Metric Screening Results 126
8.7 Literature Cited 130
8.8 Appendix E: NWCA Candidate Vegetation Metrics 133
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Chapter 9: Vegetation Multimetric Indices and Wetland Condition 166
9.1 Overview - Vegetation Multimetric Index (VMMI) 166
9.2 Calibration and Validation Data 167
9.3 Developing Vegetation Multimetric Indices (VMMIs) - Methods 168
9.3.1 Step 1 - Metric Scoring 168
9.3.2 Step 2 - Generating and Screening Candidate VMMIs 168
9.3.3 Step 3 - Determining Ecological Condition Thresholds Based on VMMI Values 169
9.4 Final 2016 VMMIs - Calculation, Performance and Condition Thresholds 170
9.4.1 VMMI Description, Metric scoring, and VMMI Calculation 171
9.4.2 VMMI Performance 174
9.4.3 Condition Thresholds for the Wetland Group VMMIs 180
9.5 Literature Cited 182
Chapter 10: Nonnative Plant Indicator (NNPI) 183
10.1 Background 183
10.2 Data Collection 184
10.3 Data Preparation 184
10.4 Nonnative Plant Indicator Overview 184
10.5 NNPI Condition Threshold Definition 185
10.6 Literature Cited 187
Chapter 11: Human-Mediated Physical Alterations 190
11.1 Data Collection 190
11.2 Scoring Each of the Six Physical Alteration Indices 193
11.3 Physical Alteration Screen Scoring (PALT_ANY and PALT_SUM) 195
11.3.1 PALT_ANY 195
11.3.2 PALT_SUM 195
11.4 Physical Alteration Stressor Condition Thresholds 196
11.5 Literature Cited 198
Chapter 12: Soil Heavy Metals 199
Chapter 13: Water Chemistry 201
13.1 Data Collection 201
13.2 Data Validation 202
13.3 Establishing a Disturbance Gradient for Sites Sampled for Water Chemistry 204
13.3.1 Development of Physical-Alterations-Possibly-Affecting-Chemicals (CALT) Indices 205
13.3.2 Screens and Thresholds for Sites Sampled for Water Chemistry 208
13.3.3 Evaluation of the Disturbance Gradient for Sites Sampled for Water Chemistry 210
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
13.4 TN and TP Stressor Condition Thresholds
212
13.5 Applying TN and TP Stressor Condition Thresholds to the NWCA 2021 Water Chemistry Data.. 214
13.6 Literature Cited 214
Chapter 14: Microcystins 217
14.1 Data Collection and Analysis 217
14.2 Application of EPA Recommended Criterion for Microcystins 217
14.3 Literature Cited 218
Chapter 15: Condition Extents, Change in Condition Extents, and Relative and Attributable Risk 221
15.1 Condition Extent Estimates 222
15.1.1 Wetland Condition Extent Estimates 223
15.1.2 Nonnative Plant Indicator (NNPI) Condition Extent Estimates 224
15.1.3 Stressor Condition Extent Estimates 225
15.2 Change and trend analysis 228
15.2.1 Data Preparation 228
15.2.2 Methods 228
15.3 Relative Extent, and Relative and Attributable Risk 228
15.3.1 Relative Extent 229
15.3.2 Relative Risk 229
15.3.2.1 Example Calculation of Relative Risk 230
15.3.2.2 Considerations When Calculating and Interpreting Relative Risk 231
15.3.2.3 Application of Relative Risk to the NWCA 231
15.3.3 Attributable Risk 232
15.3.3.1 Considerations When Interpreting Attributable Risk 233
15.4 Where to Find the Summary of NWCA Results 233
15.5 Literature Cited 233
Glossary 236
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
List of Figures
Figure 1-1. Annotated analysis flow chart indicating the chapter number (abbreviated as "CHP") in which
details may be found 24
Figure 2-1. Regions captured in the seventeen NWCA Survey Design Units 27
Figure 3-1. Example of map created using ArcGIS software to evaluate candidate handpicked sites.
Information from an assessment of the aerial imagery was recorded for basic and landscape
screening criteria 38
Figure 3-2. Map of the conterminous US showing distribution of handpicked sites (triangles) in relation
to probability sites (circles) sampled in the NWCA 2021 40
Figure 6-1. Diagram of the disturbance gradient used in the NWCA with three classes of disturbance... 53
Figure 6-2. A visual summary of how rules for assigning abiotic disturbance classes based on the physical
and chemical screens are applied to a site, where L = "least disturbed", I = "intermediate
disturbed", M = "most disturbed", and ? = "unknown". Note that the physical and chemical
screens were evaluated together to determine the abiotic disturbance class assignment for a
site 60
Figure 6-3. Map of sampled sites and their final disturbance class (REF_NWCA) assignments 63
Figure 7-1. Overview of vegetation data preparation and analysis steps used in assessing NWCA
wetlands 75
Figure 7-2. Standard NWCA Assessment Area (AA) (shaded circular area) and standard layout of
Vegetation Plots 77
Figure 7-3. Diagram of a Vegetation Plot illustrating plot boundaries and positions of nested quadrats. 77
Figure 7-4. Overview of vegetation data collection protocol for the NWCA 2021 (USEPA 2021a) 78
Figure 7-5. Process for screening and reconciling names of plant taxa observed in the NWCA. Dark blue
boxes = steps completed using R code, light blue boxes = steps requiring botanical review,
purple boxes = type of name resolution applied, and the dark blue central box = final name
resolution 84
Figure 7-6. Distribution of native status among taxon-state pairs presented as percentages 92
Figure 7-7. Text box outlining C-value selection decision tree when multiple C-values were available for
one taxon-region pair 96
Figure 8-1. Distribution of probability and hand-picked sites sampled in the 2011 and 2016 NWCA
surveys within Six Reporting Units (RPT_UNIT_6). TDL = coastal areas where tidally-
influenced estuarine wetlands occur. Inland wetlands are mapped within five geographic
regions 116
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Figure 8-2.
Six Reporting Units and four Wetland Groups: TDL = coastal areas where tidally-influenced
estuarine wetlands occur. Inland wetlands are mapped within 5 NWCA Aggregated
Ecoregions 118
Figure 8-3. Detrended correspondence analysis for NWCA 2011 and 2016 sampled sites. Sites are color-
and symbol-coded by RPT_UNIT12. Blue and TDL = Tidally-influenced, estuarine wetland
sites. Other codes and colors = Inland wetland sites by geographic region. Open symbols =
herbaceous (H) wetlands. Filled symbols = woody (W) wetlands. Note: Among the unique
1985 sampled sites, 208 were resampled sites (sampled in 2011 and 2016), (Section 6.1), and
for these resampled sites the data from 2011 visit were used in this DCA 120
Figure 9-1. Criteria for setting VMMI thresholds for good, fair, and poor condition categories based on
VMMI values observed for least-disturbed sites (REF_NWCA = L). Box plot whiskers: lower =
the 25th percentile -1.5 X IQR (interquartile range), upper = the 75th percentile + 1.5 X IQR.
170
Figure 9-2. Comparison of VMMI Estuarine Herbaceous wetlands (VMMI-EH) for least-disturbed and
most-disturbed sites. Top graph: Compares VMMI values for least- and most-disturbed EH
sites in the calibration and validation data sets. Bottom graph: VMMI values for all least- and
most-disturbed sampled EH sites. Box plots: box is interquartile (IQR) range, line in box is the
median, and whiskers represent most extreme point a distance of no more than 1.5 x IQR
from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled 176
Figure 9-3. Comparison of VMMI Estuarine Woody wetlands (VMMI-EW) for least-disturbed and most-
disturbed sites. Top graph: Compares VMMI values for least- and most-disturbed EW sites in
the calibration and validation data sets. Bottom graph: VMMI values for all least- and most-
disturbed sampled EW sites. Box plots: box is interquartile (IQR) range, line in box is the
median, and whiskers represent most extreme point a distance of no more than 1.5 x IQR
from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled 177
Figure 9-4. Comparison of VMMI Inland herbaceous wetlands (VMMI-PRLH) for least-disturbed and
most-disturbed sites. Top graph: Compares VMMI values for least- and most-disturbed PRLH
sites in the calibration and validation data sets. Bottom graph: VMMI values for all least- and
most-disturbed sampled PRLH sites. Box plots: box is interquartile (IQR) range, line in box is
the median, and whiskers represent most extreme point a distance of no more than 1.5 x
IQR from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled 178
Figure 9-5. Comparison of VMMI Inland woody wetlands (VMMI-PRLW) for least-disturbed and most-
disturbed sites. Top graph: Compares VMMI values for least- and most-disturbed PRLW sites
in the calibration and validation data sets. Bottom graph: VMMI values for all least- and
most-disturbed sampled PRLW sites. Box plots: box is interquartile (IQR) range, line in box is
the median, and whiskers represent most extreme point a distance of no more than 1.5 x
IQR from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled 179
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Figure 9-6. Comparison of VMMI values for Inland Woody wetlands (VMMI-PRLW) for least-disturbed
and most-disturbed sites by ecoregions. Top graph: VMMI-PRLW values by Five NWCA
Aggregated Ecoregions (ICP, EMU, PLN, ARW, WVM, (NWCA_EC05) see map in Figure 6-2 for
definitions) Bottom graph: VMMI-PRLW values for more mesic (OTHER) vs. more arid
(PLN_ARIDW) regional groups (OTHER = ICP, EMU, WVM; PLN_ARIDW = ARW & WVM). Box
plots: box is interquartile (IQR) range, line in box is the median, and whiskers represent most
extreme point a distance of no more than 1.5 x IQR from the box. Values beyond this
distance are outliers. Numbers below each box plot represent number of the least-disturbed
or most-disturbed sites sampled 181
Figure 11-1. The entire AA was evaluated using the P-l Form and 12 buffer plots were evaluated using
the P-2 Form 192
Figure 11-2. Physical Alteration plots for the NWCA 2021. Six Physical Alteration indices representing
Vegetation Removal, Vegetation Replacement, Water Addition/Subtraction, Water
Obstruction, Soil Hardening, Surface Modification, each with 8 metrics (i.e., checkboxes from
P-l and P-2 forms), are evaluated in the Assessment Area (AA) and in 12 Buffer Plots.
Numbers represent metric scoring (points) associated with observed metrics in the AA and
each Buffer Plot 194
Figure 13-1. Box and whisker plots showing differences between least-disturbed (blue) and most-
disturbed (red) (unique 2011 and 2016 Visit 1) sites among five regions (RPT_UNIT_5) for a)
total nitrogen (TN) and b) total phosphorus (TP). TDL = Tidal Saline, ICP = Inland Coastal
Plains, EMU = Eastern Mountains & Upper Midwest, PLN = Plains, and WST = West 211
Figure 13-2. Good stressor condition and poor stressor condition threshold-setting using the 75th and
95th percentiles of total nitrogen (TN) or total phosphorus (TP) concentrations among least-
disturbed sites sampled for water chemistry 213
Figure 15-1. The NWCA 2021 national extent estimates for wetland condition based on the Vegetation
Multimetric Indices (VMMIs). Wetland condition extent is presented for each condition
category by percent of the resource (i.e., percent of target wetland area for the Nation).
Error bars represent 95% confidence intervals as calculated by the R package spsurvey
(Dumelle et al. 2023) 224
Figure 15-2. The NWCA 2021 national extent estimates for Nonnative Plant Indicator (NNPI) condition.
NNPI condition extent is presented for each condition category by percent of the resource
(i.e., percent of target wetland area for the Nation). Error bars represent 95% confidence
intervals as calculated by the R package spsurvey (Dumelle et al. 2023) 225
Figure 15-3. The NWCA 2021 national extent estimates for 10 indicators of stressor condition. Stressor
condition extent is presented for each condition category by percent of the resource (i.e.,
percent of target wetland area for the Nation). Error bars represent 95% confidence intervals
as calculated by the R package spsurvey (Dumelle et al. 2023). Stressor abbreviations are
defined in Section 15.1.3 227
Figure 15-4. The NWCA 2021 relative extent of wetlands with stressors in poor condition, and the
relative risk and attributable risk of poor a) VMMI condition or b) NNPI condition when
stressor condition is poor as calculated by the R package spsurvey (Dumelle et al. 2023). For
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
relative risk, values below the dashed line (i.e., a relative risk < 1) signifies that there is no
association between the stressor and VMMI or NNPI condition 229
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
List of Tables
Table 2-1. Crosswalk between regions and Wetland Groups, and the Seventeen NWCA Survey Design
Units 26
Table 2-2. Sample size by survey design unit 27
Table 2-3. Number of sites expected to be sampled, reported by state and Twelve NWCA Reporting
Groups (RPTGRP_12) 29
Table 2-4. Summary of sample sizes required in the Minnesota survey design 31
Table 2-5. Wetland area (acres) in the Minnesota sample frame 32
Table 2-6. Wetland area (acres) in the NWCA sample frame reported by survey design unit and wetland
class 33
Table 2-7. Total estimated areal extents for the total target NWCA population, the sampled area extents,
and non-assessed area extents for the nation and by survey design unit and wetland class.
Results are reported as millions of acres or percent (%) of total estimated NWCA wetland
area for the nation or by survey design unit and wetland class.1 The number of sites in each
group is provided as n 34
Table 3-1. Distribution of 123 handpicked sites sampled in 2021 by Five NWCA Aggregated Ecoregions
and the NWCA Wetland Group 40
Table 5-1. Subpopulation information, including the parameter name that is used in the database, all the
potential subpopulations included in each subpopulation group, and a description of each
subpopulation group 47
Table 6-1. The number of Visit 1 (VI) probability and handpicked sites sampled in 2011 and 2016, with
their totals. Additionally, the numbers of resampled sites are reported in paratheses to
indicate that these are subtracted from the subtotals above. The total number of unique
probability and handpicked sites are reported with the final number of Index Visit sites (in
the red cell) used in the establishment of the NWCA disturbance gradient. Note that this
table does not include the 96 Visit 2 sites sampled in 2011 and 94 Visit 2 sites sampled in
2016, which are only used to calculate Signal-to-Noise ratios for some indicators/metrics (see
Chapter 8: for details) 54
Table 6-2. Indices and metrics used in NWCA 2016 to establish the disturbance gradient. Final indices
and metrics for which thresholds were created are in uppercase, bold type 55
Table 6-3. a) Least-disturbed thresholds and b) most-disturbed thresholds for the two physical alteration
screens and the number of sites that passed the screens (i.e., are considered candidate "least
disturbed" or "most disturbed") presented for Five Reporting Units (RPT_UNIT_5) 58
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
14
-------
Table 6-4.
a) Least-disturbed thresholds and b) most-disturbed thresholds for the two chemical screens
and the number of sites that passed the screens (i.e., are considered abiotic "least disturbed"
or "most disturbed") presented for Five Reporting Units (RPT_UNIT_5) 59
Table 6-5. n-sites of abiotic disturbance class assignments (REF_NWCA_ABIOTIC) reported by region
(RPT_UNIT_5) for Visit 1, Index Visit 2011 and 2016 sites 61
Table 6-6. The least-disturbed threshold for the biological screen, and the number of sites passing the
screen (and thus, are assigned final "least-disturbed" status as indicated in REF_NWCA) for
the Five Reporting Units (RPT_UNIT_5) 62
Table 6-7. n-sites within final disturbance class assignments (REF_NWCA) reported by region
(RPT_UNIT_5) for Visit 1, Index Visit 2011 and 2016 sites. Note that two sites (one from TDL
and another from ICP) were dropped due to insufficient vegetation data and assigned as
"unknown" 62
Table 7-1. Growth-habit categories, for species observed in the 2011, 2016 and 2021 NWCAs and used
in analysis, with a crosswalk to PLANTS database growth-habit designations. Capitalized
Growth-habit Category Names are used in calculation of Growth-habit metrics (see Section
8.8: Appendix E) 87
Table 7-2. Duration categories used in the NWCA analyses and a crosswalk to PLANTS database duration
designations for NWCA observed species. Capitalized Duration Category Codes are used in
calculation of Duration Metrics (see Section 8.8: Appendix E) 87
Table 7-3. Wetland Indicator Status (WIS) definitions. OBL, FACW, FAC, FACU and UPL defined by Lichvar
2016. NOL and UND defined by NWCA. These seven WIS Categories are used in calculating
Hydrophytic Status Metrics (Section 6.8: Appendix E). The Numeric Ecological Value
(ECOIND2) for each indicator status (UPL to OBL) is used in calculating indices describing the
hydrophytic status of the vegetation at each sampled site 88
Table 7-4. Wetland regions within which wetland indicator status for individual plant species are
defined, and a crosswalk between USACE codes and NWCA codes for these regions is
provided 88
Table 7-5. Definition of state-level native status designations for NWCA taxon-state pairs 91
Table 8-1. Numbers of unique NWCA 2011 and 2016 sampled sites. NWCA_REF (Disturbance): L = Least,
I = Intermediate, M = Most, ? = Undetermined. Revisit = site sampled twice in same field
season 116
Table 8-2. Numbers of unique NWCA 2011 and NWCA 2016 sampled sites by RPT_UNIT12 (RPT_UNIT_6
x WETCLS_GRP). RPT_UNIT_6 is defined in Figure 6-2 and Table 6-3. WETCLS_GRP is defined
in Table 6-4. REF_NWCA (Disturbance): L = Least, I = Intermediate, M = Most, ? =
undetermined. Revisit = site sampled twice in same field season 117
Table 8-3. Numbers of unique NWCA 2011 and 2016 sampled sites by six reporting units (RPT_UNIT_6).
REF_NWCA (Disturbance): L = Least, I = Intermediate, M = Most, ? = undetermined. Revisit =
site sampled twice in same field season. Tidal (TDL) = tidally-influenced estuarine wetlands
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
occurring in near coastal areas. The other five groups represent inland wetlands within five
ecoregional areas. See Table 8-4 for description of include wetland types 119
Table 8-4. Numbers of unique NWCA 2011 and 2016 sampled sites by Wetland Groups (WETCLS_GRP).
REF_NWCA (Disturbance): L = Least, I = Intermediate, M = Most, ? = undetermined. Revisit =
site sampled twice in same field season. EH and EW are tidally-influenced estuarine
wetlands. PRLH and PRLW are inland wetlands 119
Table 8-5. Metric Groups and component Metric Types for characterizing vegetation condition 123
Table 8-6. Metrics (n = 40) that passed screening criteria for the Estuarine Herbaceous (EH) wetland
subpopulation. Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section
8.8 (Appendix E) 126
Table 8-7. Metrics (n = 21) that passed screening criteria for the Estuarine Woody (EW) wetland
subpopulation. Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section
8.8 (Appendix E) 127
Table 8-8. Metrics (n = 42) that passed screening criteria for the Inland Herbaceous (PRLH) wetland
subpopulation. Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section
8.8 (Appendix E) 128
Table 8-9. Metrics (n = 47) that passed screening criteria for the Inland Woody (PRLW) wetland
subpopulation. Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section
8.8 (Appendix E) 129
Table 9-1. Metrics included in each of the four NWCA 2016 Vegetation Multimetric Indices (VMMIs). See
Section 8.8:, Appendix E for formulas for calculation of these metrics 172
Table 9-2. VMMI-EH metrics: floor and ceiling values, disturbance response, and interpolation formula
for scoring individual metrics. Final scores for each metric decrease with disturbance 172
Table 9-3. VMMI-EW metrics: floor and ceiling values, disturbance response, and interpolation formula
for scoring individual metrics. Final scores for each metric decrease with disturbance 173
Table 9-4. VMMI-PRLH metrics: floor and ceiling values, disturbance response, and interpolation formula
for scoring individual metrics. Final scores for each metric decrease with disturbance 173
Table 9-5. VMMI-PRLW metrics: floor and ceiling values, disturbance response, and interpolation
formula for scoring individual metrics. Final scores for each metric decrease with
disturbance 174
Table 9-6. Summary statistics for the final four VMMIs: EH - Estuarine Herbaceous, EW -Estuarine
Woody, PRLH - Inland Herbaceous, PRLW - Inland Woody. Statistics calculated based on
VMMI values for sampled sites and revisit sites from the calibration data set for the relevant
VMMI group 175
Table 9-7. VMMI value thresholds indicating good, fair, and poor ecological condition based on least-
disturbed sites in each Wetland Group (WETCLS_GRP). Sites with VMMI values from the 5th
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
up to the 25th percentile for least-disturbed (REF_NWCA) sites are considered in fair
condition 180
Table 9-8. Two-sample unequal variances t-tests comparing VMMI value means for all sampled least-
and most-disturbed sites for each Wetland Group VMMI 180
Table 10-1. Definition of metrics used in the NNPI 184
Table 10-2. Condition Threshold Exceedance Values for each of the metrics informing the Nonnative
Plant Indicator (NNPI): Relative Cover of Nonnative Species (XRCOV_AC), Nonnative Richness
(TOTN_AC), and Relative Frequency of Nonnative Species (RFREQ_AC) 186
Table 11-1. Six indices of human-mediated physical alterations and the 48 metrics collected on the P-l
Physical Alteration Assessment Area and P-2 Physical Alteration Buffer Plots Forms 191
Table 13-1. Water chemistry analytes measured in the laboratory, with their associated units and a
summary of methods 202
Table 13-2. Variability and repeatability of water chemistry analytes measured in the 2016 NWCA,
including below-detection rates for all 2016 NWCA sites (Visit 1 and Visit 2, probability and
handpicked), cross-visit correlations based on the 61 revisit sites, and Signal-to-Noise ratios
(S:N) for all sites and inland (freshwater) sites 204
Table 13-3. The number of Visit 1 (VI) probability and handpicked sites sampled for water chemistry in
2011 and 2016, with their totals. Additionally, the numbers of resampled sites with water
chemistry data are reported in paratheses to indicate that these are subtracted from the
subtotals above. The total number of unique probability and handpicked sites with water
chemistry data are reported with the final number of Index Visit sites (in the red cell) used in
the establishment of the water chemistry disturbance gradient. Note that this table does not
include the 51 Visit 2 sites with water chemistry sampled in 2011 and 64 Visit 2 sites with
water chemistry sampled in 2016, which are only used to calculate Signal-to-Noise ratios. 205
Table 13-4. Subset of Physical Alteration metrics (defined in Chapter 11:, Section 11.2) assigned to the
Physical-Alterations-Possibly-Affecting-Nutrient (CALT_NUT), Physical-Alterations-Possibly-
Affecting-Suspended Sediments (CALT_SED), and Physical-Alterations-Possibly-Affecting-
Salinity (CALT_SAL) indices. "X" indicates that the Physical Alteration metric was included the
CALT index. Note that not all Physical Alteration metrics were assigned to a CALT index.
Write-in "others" from the H-l and B-l Forms were not considered and are therefore
excluded from the list of Form Items 206
Table 13-5. Six water chemistry screens and their least-disturbed and most-disturbed thresholds used to
assign disturbance class to each site sampled for water chemistry 209
Table 13-6. n-sites sampled for water chemistry, presented by disturbance class assignments
(unpublished) reported by region (RPT_UNIT_5) for Visit 1, Index Visit 2011 and 2016 sites.
210
Table 13-7. Final total nitrogen (TN) and total phosphorus (TP) thresholds and relevant information for
developing those thresholds, including the number of least-disturbed sites with water
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
chemistry on which threshold percentiles are based (see Section 13.3 for details), the high
outlier cut-off, and the number of outlier sites 213
Table 15-1. Example contingency table for relative risk that reports the proportion of stream length
associated with good and poor condition (as indicated by Fish Index of Biotic Integrity, IBI)
and good and poor stressor condition (as indicated by stream water total nitrogen
concentration, TN). Results are hypothetical 230
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Acronym List
AA
Assessment Area
AR
Attributable Risk
BPJ
Best Professional Judgement
CCs
Coefficients of Conservatism
CDF
Cumulative Distribution Function
C-value
Coefficients of Conservatism
EF
Enrichment Factor
FQAI
Floristic Quality Assessment Index
GIS
Geographic Information System
GRTS
Generalized Random Tessellation Stratified (in relation to the survey design)
HGM
Hydrogeomorphic Class
HMI
Heavy Metal Index
IM
Information Management
IQR
Interquartile Ranges
MDL
Minimum Detection Limit
Mean C
Mean Coefficients of Conservatism
NARS
USEPA National Aquatic Resource Surveys
NFQD
National Floristic Quality Database
NPS
US National Park Service
NNPI
Nonnative Plant Indicator
NRCS
Natural Resources Conservation Service
NWCA
USEPA National Wetland Condition Assessment
NWPL
National Wetland Plant List
ORD
USEPA Office of Research and Development
OW
USEPA Office of Water
PQL
Practical Quantitation Limit
OA
Quality Assurance
RR
Relative Risk
S&T
USFWS Status and Trends
S:N
SignaLNoise (i.e., signal to noise ratio)
UID
Unique Identification
USACE
United States Army Corps of Engineers
USDA
United States Department of Agriculture
USEPA
United States Environmental Protection Agency
USFWS
United States Fish and Wildlife Service
VMMI
Vegetation Multimetric Index
WIS
Wetland Indicator Status
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
19
-------
This page was intentionally left blank for double-sided printing.
National Wetland Condition Assessment: 2021 Technical Support Document
Dec 2024
-------
Chapter 1: Project Overview and Elements of Analysis
1.1 Overview
The National Wetland Condition Assessment (NWCA) is a collaboration among the U.S. Environmental
Protection Agency (USEPA), States, Tribes, and other partners. It is part of the National Aquatic Resource
Surveys (NARS) program to conduct national scale assessments of aquatic resources. The NWCA 2021
provides condition assessment results at national and regional scales of the ecological condition of
wetlands. This assessment was accomplished by collecting and analyzing biological, chemical and physical
data from sampled wetlands across the conterminous United States.
This document, the National Wetland Condition Assessment: 2021 Technical Support Document,
accompanies the National Wetland Condition Assessment: The Third Collaborative Survey of Wetlands in
the United States (referred to as the "Web Report"). The Web Report describes the background and main
findings of the NWCA 2021. The Technical Support Document supports the findings presented in the
Public Report by describing the development of the survey design and the scientific methods used to
collect, evaluate, and analyze data collected for the NWCA 2021.
The Technical Support Document includes information on the target population, sample frame, and site
selection underlying the NWCA 2021 survey design. The report provides a synthesis of data preparation
and management processes, including field and laboratory data entry, review, and several quality
assurance checks used in analysis. The NWCA evaluates the condition of and potential stressors to
wetlands along a gradient of disturbance, based on comparison to sites designated as "least-disturbed"
(or "reference"). The Technical Support Document provides a thorough overview of the development and
application of this approach.
1.2 Objectives of the NWCA
The objective of the NWCA is to characterize aspects of the biological, chemical and physical condition of
wetlands throughout the conterminous United States. It employs a statistically valid probability design
stratified to allow estimates of the condition of wetlands at national and regional scales.
The NWCA is designed to answer the following questions about wetlands across the United States.
1. What is the current biological, chemical, and physical condition of wetlands?
a. What is the extent of stressors among wetlands?
b. Are stressors widespread (e.g. national) or localized (e.g. regional)?
2. Is the proportion of wetland area in poor condition getting better, worse or staying the same over
time?
3. Which stressors are most strongly associated with degraded biological condition in wetlands?
A variety of biological, chemical, and physical data were collected and developed into several indicators of
ecological condition or stress to wetlands that inform the population estimate results of the 2016 NWCA.
For each of these indicators the Technical Support Document provides background and underlying
rationale, evaluation of candidates, and development of the final indicators chosen for the NWCA,
including defining threshold categories for condition and disturbance in order to evaluate and compare
data.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
The information described in the Technical Support Document was developed through the efforts and
cooperation of NWCA scientists from USEPA, technical experts and participating cooperators from
academia and state and tribal wetland programs. While this Technical Support Document provides a
comprehensive summary of NWCA procedures, including design, sampling, and analysis of data, it is not
intended to present an in-depth report of data analysis results.
1.3 Elements of Analysis
The analysis for the NWCA 2021 involved a number of interrelated tasks composed of multiple steps. This
brief overview of the entire process provides a context for the details of each of the major tasks described
in this document.
Figure 1-1, which can be found on the following page, illustrates the analysis process, beginning with field
sampling probability and handpicked sites (left side of chart) and concluding with the population
estimates of wetland condition extent, stressor condition extent, and relative and attributable risk (right
side of chart) for the NWCA target population in the conterminous US. The components of each of the
major tasks are indicated by text boxes and include the chapter number in which details may be found.
The key elements of the analysis outlined in the flowchart are:
1) Field sampling using protocol from the 2021 Field Operations Manual (USEPA 2021) results in
data acquisition for probability (Chapter 2:) and handpicked sites (Chapter 3:). This data is
prepared, and quality assurance continues throughout all of the analyses resulting in the
production of the data tables used by the analysts (Chapter 4:).
2) Metrics and indices used to develop disturbance and stressor thresholds are calculated for each
site, including Human-Mediated Physical Disturbances (Chapter 11:), Soil Heavy Metals (Chapter
12:), Percent Relative Cover of Nonnative Plant Species (Section 6.6), Nonnative Plant Index
(NNPI) (Chapter 10:), Water Chemistry (Chapter 13:), and Microcystins (Chapter 14:).
3) Three types of data are used to develop disturbance gradient thresholds and categorize each site
as least-, intermediate-, or most-disturbed (Chapter 6:). These data types include physical data
(human-mediated physical alterations), chemical data (soil heavy metals), and biological data
(percent relative cover of nonnative plant species).
4) Five types of data are used to develop stressor thresholds, including human-mediated physical
alterations, soil heavy metals, Nonnative Plant Index (NNPI), water chemistry, and microcystins,
found at the end of their individual chapters (Chapter 10: through Chapter 14:).
5) To develop Vegetation Multimetric Indices (VMMIs), first a vegetation analysis approach is
identified and data acquisition and preparation (Chapter 7:) is conducted, followed by
prerequisite analyses to vegetation indicator development (Chapter 8:). Using least- and most-
disturbed sites, VMMIs are developed and thresholds for good, fair, and poor wetland condition
are established (Chapter 9:).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
6) Finally, site weights and only probability sites are used to calculate results for the wetland
population (Chapter 15:) and various subpopulations (Chapter 5:). Results include wetland
condition extent, stressor condition extent, change in both wetland condition extent and stressor
condition extent from the 2011, 2016 and the NWCA 2021, and relative and attributable risk.
Final results are published using the online dashboard at
https://wetlandassessment.epa.gov/dashboard.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
REPORTING (CHP 15)
and NWCA Public Reporting
DATA
PREPARATION
(CHP 4)
prepare & QA
data before use
SURVEY
DESIGN
(CHP 2)
Probability
sites (PROB)
HANDPICKED
SITES
(CHP 3)
handpicked
sites (HAND)
*
collect data
in the field
using protocol
from the Field
Operations
Manual
for each site, calculate metrics and indices,
including:
NONNATIVE PLANT
INDICATOR (NNPI) (CHP 10)
WATER CHEMISTRY (CHP 13)
MICROCYSTES (CHP 14)
HUMAN-MEDIATED
PHYSICAL ALTERATIONS (CHP 11)1
SOIL HEAVY METALS (CHP 1211-
PERCENT RELATIVE COVER OF
NONNATIVE PLANTS (SECT 6.6)
STRESSOR
THRESHOLDS
(CHP 10-14)
DISTURBANCE
THRESHOLDS
(CHP 6)
least disturbed i
intermediate
disturbed
most disturbed I
H
develop vegetation
multimetric indices
VEGETATION
ANALYSIS
OVERVIEW-,
DATA
ACQUISITION,
PREP (CHP 7)
~
PREREQUISITE
ANALYSES TO
VEGETATION
INDICATOR
DEVELOPMENT
(CHP 8)
~
¦ M VEGETATION
I MULTIMETRIC
INDICES &
WETLAND
I r CONDITION
¦ J (CHP 9)
©
r * _ site weights from
i VV g " probabiliy design
only probability sites
used for population
estimates:
STRESSOR
CONDITION
EXTENT & CHANGE
good
fair
poor
very poor
t:
NNPI only
RELATIVE &
ATTRIBUTABLE
RISK
WETLAND
CONDITION
EXTENT & CHANGE
good
fair
poor
SUBPOPULATIONS
(CHP 5)
Figure 1-1, Annotated analysis flow chart indicating the chapter number (abbreviated as "CHP") in which details may be found.
24
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Chapter 2: Survey Design
The NWCA was designed to assess the ecological condition of broad groups or populations of wetlands,
rather than as individual wetlands or wetlands across individual states. The NWCA design allows
characterization of wetlands at national and regional scales using indicators of ecological condition and
stress. It is not intended to represent the condition of individual wetlands. The statistical design also
accounts for the distribution of wetlands across the country - some areas have fewer wetlands than
others - so that, even in areas of the country where there are few sample sites, regional and national
results still apply to the broader target population.
2.1 Description of the NWCA Wetland Type Population
The target population for the NWCA included all wetlands of the conterminous United States not
currently in crop production, including tidal and nontidal wetted areas with rooted vegetation and, when
present, shallow open water less than one meter in depth. A wetland's status under state or federal
regulatory programs did not factor into this definition. Wetland attributes are assumed to vary
continuously across a wetland.
2.2 Sample Frame, Survey Design, and Site Selection
Probability sites that were sampled as part of the NWCA were selected using a sampling frame on which
the survey design was based. The following sections provide details about how the sample frame and
survey design were developed, and how sites were selected.
2.2.1 Sampling frame
The foundation of the survey design is a sampling frame, i.e., the geographic data layer that identifies
locations and boundaries of all wetlands that meet the definition of the target population. The sampling
frame source is the U.S. Fish & Wildlife Service's National Wetland Inventory (NWI). The NWI wetland
polygon data was updated on October 8, 2019 and is the latest available. To obtain the sampling frame,
NWCA processed the data by assigning wetland polygons to states and within each state assigning them
to the NARS nine aggregated ecoregions. In addition, the detailed wetland types were categorized into
seven wetland types of interest to NWCA (E2EM, E2SS, PEM, PSS, PFO, Pf and PUBPAB) and five wetlands
types not included (EOTH - estuarine other wetlands, M1M2 - marine wetlands, LOTH - lacustrine other
wetlands, POTH - palustrine other wetlands, and ROTH - riverine other wetlands). The former are
included as they are likely to result in sites that would meet the NWCA definition of a wetland and the
latter are excluded as they are unlikely to result in sites that would meet the NWCA definition of a
wetland. Cowardian wetland classes were assigned to each NWCA wetland class by two wetland
ecologists. Two exceptions to this were Montana and Minnesota. Montana provided a GIS layer similar
to NWI that had not yet been incorporated. Minnesota conducts a wetland quantity survey
similar to U.S. Fish & Wildlife Service's Status and Trends program. It is based on 1 sq mi plots.
NWCA used the results of this survey for the Minnesota sampling frame. The Minnesota survey
design also differs from the NWCA survey design completed for other states. In summary, the
wetland types incldued are:
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
E2EM - estuarine intertidal emergent (estuarine herbaceaous)
E2SS - estuarine intertidal scrub shrub/forested (estuarine woody)
PEM - palustrine emergent (inland herbaceous)
PSS - palustrine scrub shrub (inland woody)
PFO - palustrine forested (inland woody)
Pf-palustrine farmed (inland herbaceous)
PUBPAB - palustrine unconsolidated bottom/aquatic bed (inland herbaceous)
2.2.2 Survey design units
NWCA 2021 is based on 17 survey design units associated with 10 geographic areas. Inland wetland
regions are based on NARS nine aggregated ecoregions where the Upper Midwest and Northern
Appalachians are combined to form the North Central East region and the Northern Plains and Southern
Plains are combined to form the Great Plains region. Estuarine wetland coastal regions are the Pacific
Coast, Gulf and Florida Coast and the Atlantic Coast. The inland survey design regions are then divided by
herbaceous and woody wetland types.
Table 2-1. Crosswalk between regions and Wetland Groups, and the Seventeen NWCA Survey Design Units.
Region Description
Survey Design UnitDescription
Survey Unit Code
Atlantic Estuarine
Estuarine
ATL
Gulf Estuarine
Estuarine
GFL
Pacific Estuarine
Estuarine
PAC
Coastal Plains (CPL)
Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
CPL-PRLH
Palustrine, Riverine, and Lacustrine Woody (PRLW)
CPL-PRLW
North-Central East (NCE)
Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
NCE-PRLH
Palustrine, Riverine, and Lacustrine Woody (PRLW)
NCE-PRLW
Southern Appalachians (SAP)
Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
SAP-PRLH
Palustrine, Riverine, and Lacustrine Woody (PRLW)
SAP-PRLW
Temperate Plains (TPL)
Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
TPL-PRLH
Palustrine, Riverine, and Lacustrine Woody (PRLW)
TPL-PRLW
Great Plains (GPL)
Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
GPL-PRLH
Palustrine, Riverine, and Lacustrine Woody (PRLW)
GPL-PRLW
Western Mountains (WMT)
Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
WMT-PRLH
Palustrine, Riverine, and Lacustrine Woody (PRLW)
WMT-PRLW
Xeric West (XER)
Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
XER-PRLH
Palustrine, Riverine, and Lacustrine Woody (PRLW)
XER-PRLW
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
26
-------
2021 National Wetland Condition Assessment Design
Tidal Saline
H Atlantic Coast (ATL)
Gulf and Florida Coasts (GFL)
| Pacific Coast (PAC)
Figure 2-1. Regions captured in the seventeen NWCA Survey Design Units.
2.2.3 Expected Sample Size
The expected sample size is 904 sites for conterminous 48 states with 96 of those sites to be revisited
twice for a total of 1,000 site-visits. Each state will have two sites that will be visited twice in 2021 for a
total of 1,000 site-visits. Sample allocation depends on the 17 reporting units with 53 in each unit
except for three (3) additional sites in CPL woody for total of 904 sites. The number of unique sites for
each state is proportional to the wetland area in each survey design unit with a restriction that each
state has a minimum of eight (8) unique sites so that with the two revisits a minimum of 10 site visits in
each state.
Table 2-2. Sample size by survey design unit
Survey Design Unit
Total # Sites
PRLH Sites
PRLW Sites
NCE
106
53
53
SAP
106
53
53
CPL
112
53
56
TPL
106
53
53
GPL
106
53
53
WMT
106
53
53
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
XER
106
53
53
ATL
53
GFL
53
PAC
53
Total
904
371
374
2.2.4 Survey design
The NWCA 2021 survey design consists of two main components: sites from the prior NWCA 2016 suvey
(re-sampled sites) and new sites for 2021.
2.2.4.1 Re-sampled Site Design
Approximately 30 percent of the 904 sites, 269 sites, were selected to be re-sampled from NWCA 2016.
The actual number of sites was determined for each state after the number of sites in each state was
determined based on wetland area. Sites were restricted to NWCA 2016 evaluated sites and then
ordered by state and 2016 sitelD within state. The first "n" sites within a state were designated
Base21_16RVT2 (first 2 sites), Base21_16 and the remaining sites as Over21_16. This means that some
of the Base21_16 sites will likely not have been target and sampled in 2016. Hence over sample sites will
likely be required. Note that all sites were evaluated again in 2021 to determine if they are target
population and if in the target population whether they can be sampled.
2.2.4.2 New Site Design
The remaining approximately 70%, 635 sites, are new sites. The new site survey design is initially
stratified by state and reporting unit to ensure sample size requirements are met. An over sample 20
times the number of sites required for the reporting unit was included. Sites were selected using a
Generalized Random Tessellation Stratified (GRTS) survey design for an area resource. The NWCA uses a
GRTS survey design to select sites, which provides spatially distributed samples, and thus, are more likely
to be representative of the population than other common spatial survey designs (Stevens and Olsen
2004, Olsen et al. 2012). After the sites were selected, a second survey design was completed on the
selected sites for the purpose of removing the reporting unit stratification so that sites could be
replaced within a state by inland wetland and, if present, estuarine wetland. Site selection was
completed using the R package 'spsurvey' (Kincaid and Olsen 2019). To select sites using the survey
design, five panels were included from which set points[i.e., site coordinates selected by the survey
design) were to be sampled in the listed order (USEPA 2021a, b). The panels (in order) were:
Base21_16RVT2: identifies sites from NWCA 2016 that are to be visited twice within the 2021 season
(i.e., both a resample and a revisit site),
Base21_16: identifies sites from NWCA 2016 that are to be visited once within the 2021 season.
Over21_16: identifies sites from NWCA 2016 to be used if one or more Base21_16 sites cannot be
sampled.
Base21_21: identifies new sites to be visited once.
Over21_21: identifies new sites to be used if one or more Base21_21 sites cannot be sampled.
The sites were ordered in reverse hierarchical order to ensure that the final set of sites evaluated satisfied
the requirements for a probability survey design (Stevens and Olsen 2004). Sites were sampled based on
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
this order. All sites - from the first one on the list through the last site sampled from the list - were
evaluated and, hence, included in the study.
Sites that were originally sampled in the field in the previous NWCA survey (i.e., 2016) and selected to be
sampled again in the current survey (i.e., 2021). The resample design included 269 sites sampled in the
2016 NWCA. The actual number of sites was determined for each state after the number of sites in each
state was determined based on wetland area. Sites were restricted to NWCA 2016 evaluated sites and
then ordered by state and 2016 sitelD within state. The first "n" sites within a state were designated
Base21_16RVT2 (first 2 sites), Base21_16 and the remaining sites as Over21_16. This means that some of
the Base21_16 sites will likely not have been target and sampled in 2016. Hence over sample sites will
likely be required. Note that all sites should be evaluated again in 2021 to determine if they are target
and if target whether they can be sampled. Conditions could have changed since 2016.
2.2.5 Number of Sites Expected to be Sampled
The expected sample size was 904 probability sites for the conterminous 48 states made up of 269
resampled sites from NWCA 2016 and 635 new probability sites. Each state was expected to revisit two
sites within the field season, adding 96 revisits. Therefore, 1,000 site visits (i.e., sampling events) were
expected for the NWCA 2021. The minimum expected number of sites to be sampled in a state was eight,
with two of these sites revisited, for a total often site visits. The maximum number of sites for a state was
74 (California) (Table 2-3). Additional sites were sampled in some states with the objective of enabling a
state-level assessment.
Table 2-3. Number of sites expected to be sampled, reported by state and Twelve NWCA Reporting Groups
RPTGRP_12).
STATE
EST
PRLH
PRLW
Total
NEW_21
RVST16
NWCA16
AL
1
4
11
16
11
5
12
AR
0
5
5
10
7
3
10
AZ
0
6
9
15
11
4
13
CA
36
18
20
74
52
22
40
CO
0
11
11
22
15
7
23
CT
2
2
4
8
6
2
4
DE
2
2
4
8
6
2
10
FL
18
20
10
48
34
14
51
GA
10
5
12
27
19
8
19
IA
0
5
5
10
7
3
7
ID
0
7
8
15
11
4
30
IL
0
4
11
15
11
4
18
IN
0
4
8
12
8
4
7
KS
0
5
3
8
6
2
6
KY
0
6
6
12
8
4
44
LA
27
8
8
43
30
13
37
MA
2
2
4
8
6
2
10
MD
6
2
2
10
7
3
8
ME
2
2
4
8
6
2
4
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Ml
0
6
14
20
14
6
12
MN
0
28
13
41
29
12
13
MO
0
11
11
22
15
7
13
MS
1
3
5
9
6
3
13
MT
0
15
10
25
17
8
19
NC
6
4
9
19
13
6
11
ND
0
20
2
22
15
7
11
NE
0
7
6
13
9
4
5
NH
2
3
3
8
6
2
5
NJ
5
2
3
10
7
3
12
NM
0
6
4
10
7
3
11
NV
0
11
14
25
17
8
17
NY
1
4
3
8
6
2
11
OH
0
4
4
8
6
2
5
OK
0
9
24
33
23
10
13
OR
6
15
11
32
22
10
42
PA
0
4
4
8
6
2
5
Rl
2
2
4
8
6
2
8
SC
9
4
6
19
13
6
15
SD
0
17
3
20
14
6
15
TN
0
4
4
8
6
2
4
TX
6
18
21
45
31
14
33
UT
0
14
3
17
12
5
22
VA
4
5
6
15
11
4
10
VT
0
3
5
8
6
2
4
WA
11
8
13
32
22
10
34
Wl
0
10
13
23
16
7
13
WV
0
6
2
8
6
2
2
WY
0
10
9
19
13
6
23
Total
159
371
374
904
635
269
754
2.2.6 State-Requested Modifications to the Survey Design
Minnesota elected to modify the survey design for their state because of the availability of additional
wetland mapping information. In 2006, Minnesota developed a Comprehensive Wetland Assessment,
Monitoring, and Mapping Strategy (CWAMMS). One of the primary outcomes of the CWAMMS was the
development of statewide random surveys under the Wetland Status and Trends Monitoring Program
(WSTMP), to begin assessingthe status and trends of wetland quantity and quality in Minnesota (Kloiber
2010). The wetland quantity survey, implemented by the Minnesota Department of Natural Resources,
was modeled after the USFWS S&T program (Dahl 2006, 2011). The WSTMP survey design was the basis
for the Minnesota NWCA design.
The WSTMP design contains l-mi2 grid cells for Minnesota (and requires that at least 25% of grid cell be
within state of Minnesota) where the grid matches the USFWS S&T 4-mi2 grid boundaries. Each 4-mi2
grid cell was subdivided into four l-mi2 grid cells. An equal-probability GRTS survey design was used to
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
select 4,740 l-mi2 plots. All wetland habitats within these plots were delineated using aerial imagery
obtained in years 2009, 2010, and 2011. Where portions of some l-mi2 plots fell outside of state
boundaries, only the portion occurring within the state was photo-interpreted and mapped. Therefore,
the total area of the sample frame extent was less than 4,740 mi2. The following wetland classes were
defined to meet NWCA wetland classes:
PFO = cf'FO")
PSS = c("SS")
PEM = cf'EM", "EMm")
PUBPAB = c(MAB", "ABm", "UB", "UBm")
Pf = cf'CW")
The survey design must satisfy the sample size requirements for NWCA 2021 and the 2021 Minnesota
intensification. NWCA requires a total of 41 sites where 12 sites are resampled sites from 2016 and 29
are new sites for 2021. Expectation for NWCA is that 28 sites would be from Upper Midwest aggregated
ecoregion and 13 from Temperate Plains aggregated ecoregion. Minnesota requires 150 sites with 50
sites each in Mixed Wood Plains, Mixed Wood Shield and Temperate Prairies ecoregions (Table 2-3).
Table 2-4. Summary of sample sizes required in the Minnesota survey design
Ecoregion
Organization
New 2021
Resample 2016
Resample 2011
Total
Mixed Wood Plains
NWCA + MN
10
4
0
14
MN Only
15
9
12
36
Total
25
13
12
50
Mixed Wood Shield
NWCA + MN
10
4
0
14
MN Only
15
9
12
36
Total
25
13
12
50
Temperate Plains
NWCA + MN
9
4
0
13
MN Only
16
9
12
37
Total
25
13
12
50
NWCA Total
29
12
0
41
MN Only Total
75
39
36
150
The Minnesota survey design has two main components: selection of sites from prior surveys and
selection of new sites.
Prior Survey Site Design. Minnesota provided an Excel file with all sites evaluated for Minnesota in 2016,
including those not needed. All not-needed sites were eliminated and then the sites were sorted by
Minnesota ecoregion, then PANEL16 and then EPASITEID16. Sites were assigned a stratum based on
ecoregion and then within an ecoregion a PANELJJSE variable was created. Within the stratum sites
were assigned in sitelD order to the panels Base21_16_MN_NWCA_RVT2, Base21_16_MN_NWCA,
Base21_16_MN and Base21_ll_MN to meet the sample size requirements. All remaining sites within a
stratum were assigned to panels Over21_16 or Over21_ll. Note that the process ignores whether the
site was evaluated as nontarget, target not sampled or target sampled. Consequently, it is expected that
over sample sites will be required.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
The new site survey design is stratified by Minnesota ecoregion with unequal probability of selection by
herbaceous and woody wetland types. Each ecoregion has a sample size of 25 sites assigned to two
panels in PANEL_USE: Base21_21_MN_NWCA and Base21_21_MN to meet sample size requirements in
Table 2. Additional sites are in the panel Over21_21. The new survey design has six base panels:
Base21_16_MN_NWCA_RVT2 - Panel of sites sampled in 2016 to be sampled twice for NWCA
2021. Sites are for NWCA as well as Minnesota intensification.
Base21_16_MN_NWCA- Panel of sites sampled in 2016 to be sampled once for NWCA 2021.
Sites are for NWCA as well as Minnesota intensification.
Base21_21_MN_NWCA - Panel of new sites to be sampled in 2021. Sites to be sampled for
NWCA as well as Minnesota intensification.
Base21_ll_MN - Panel of sites sampled in 2011 and 2016 to be sampled again in 2021 for
Minnesota intensification.
Base21_16_MN - Panel of sites sampled in 2016 to be sampled again in 2021 for Minnesota
intensification.
Base21_21_MN - Panel of new sites to be sampled in 2021 for Minnesota intensification.
The new survey design has three over sample panels:
Over21_ll - Additional sites to be evaluated when Base21_ll sites are not available.
Over21_16 - Additional sites to be evaluated when Base21_16 sites are not available.
Over21_21 - Additional sites to be evaluated when Base21_21 sites are not available.
The Minnesota sample frame summary of 4,740 1-square mile plots (portion within the state) is given by
their cover code and the three Minnesota ecoregions.
Table 2-5. Wetland area (acres) in the Minnesota sample frame.
Cover Code
MWP
MWS
TPL
Total
PEM
54,697
87,630
35,912
178,239
Pf
2,536
1,056
5,489
9,080
PFO
15,622
210,648
8,962
235,232
PSS
18,806
104,757
8,942
132,505
PUBPAB
16,938
15,451
7,698
40,088
Excluded Wetlands
26,421
99,242
11,739
137,402
Upland
595,043
693,918
995,609
228,4569
Total
730,063
1,212,702
1,074,350
3,017,115
2.3 Wetland Area in the NWCA Sample Frame
119,808,811 acres are included in the NWCA sample frame. The wetland area included in the NWCA 2021
sample frame is provided in Table 2 6 summarized by reporting domain.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
32
-------
Table 2-6. Wetland area (acres
in the NWCA sample frame reported by survey design unit and wetland class.
Region
EOTH
EST
LOTH
M1M2
POTH
PRLH
PRLW
ROTH
Total
ATL
1,084,891
1,592,982
0
68,681
0
0
0
0
2,746,554
CPL
0
0
3,311,394
0
35,946
9,214,653
34,136,743
1,784,049
48,482,785
GFL
2,482,748
2,690,574
0
61,025
0
0
0
0
,5234,348
GPL
0
0
2,028,600
0
3,970
8,753,988
1,073,167
457,102
12,316,826
NCE
0
0
6,092,439
0
23,789
6,444,900
20,269,081
1,060,699
33,890,907
PAC
125,718
80,424
0
27,087
0
0
0
0
233,230
SAP
2
0
2,452,141
0
6,519
2,480,256
2,788,725
1,746,543
9,474,186
TPL
0
0
1,615,596
0
28,906
7,367,449
3,802,818
971,326
13,786,093
WMT
0
0
1,902,575
0
1,8521
5,469,099
1,355,735
487,878
9,233,808
XER
0
0
5,114,093
0
28,770
11,077,044
1,211,176
317,677
17,748,759
Total
3,693,360
4,363,980
22,516,838
156,793
146,421
50,807,387
64,637,444
6,825,273
153,147,497
2.4 Survey Analysis
Any statistical analysis of data must incorporate information about the monitoring survey design. When
estimates of characteristics for the entire target population are computed, called population estimates
(discussed in Chapter 15), the statistical analysis must account for any stratification or unequal probability
selection in the design. The statistical analysis of the NWCA population estimates were completed using
the R package 'spsurvey' (Dumelle et al. 2023), which implements the methods described by Diaz-Ramos
etal. (1996).
2.5 Estimated Wetland Extent of the NWCA Wetland Population and
Implications for Reporting
Using a site evaluation process (USEPA 2021b), points selected by the NWCA survey design were
screened using aerial photo interpretations and GIS analyses to eliminate locations not suitable for NWCA
sampling (e.g., non-NWCA wetland types, wetlands converted to non-wetland land). Sites could also be
eliminated duringfield reconnaissance if they were a non-target type or could not be assessed due to
accessibility issues. Dropped sites were systematically replaced from a pool of replacement sites (i.e.,
oversample panel discussed in Section 2.2.3) from the survey design.
Eliminated sites affect how the final population results are estimated and reported. Accounting for non-
NWCA wetland types (e.g., wetlands in active crop production, deeper water ponds, mudflats), there
were an estimated 81.7 million acres of wetlands in the population across the conterminous US.
Throughout this report, wetland area as percentages is relative to the 81.7 million acres.
Table 2-7 illustrates the distribution of estimated extents of the 1) total NWCA wetland population, 2) the
sampled area (based on sampled probability sites), and 3) non-assessed area (based on probability sites
that could not be assessed) for the nation (conterminous US) and survey design unit and wetland class.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
33
-------
Table 2-7. Total estimated areal extents for the total target NWCA population, the sampled area extents, and non-
assessed area extents for the nation and by survey design unit and wetland class. Results are reported as millions
of acres or percent (%) of total estimated NWCA wetland area for the nation or by survey design unit and wetland
class.1 The number of sites in each group is provided as n.
Target NWCA
Other
Wetland
Sampled
Access Denied
Inaccessible
Non-Assessed
Population
millions acres
millions acres
millions acres
millions acres
Region
millions acres
(% area)
(% area)
(% area)
(% area)
Nation
81.7
34.3 (42%)
35.7 (44%)
6.1 (8%)
5.5 (7%)
n = 933
n = 1132
n = 239
n = 235
ATL
1.1
0.9 (84%)
0.1 (7%)
0.1 (6%)
<0.1 (3%)
n =57
n = 8
n = 3
n = 3
GFL
2.0
0.7 (33%)
0.6 (31%)
0.4(18%)
0.4 (18%)
n =57
n = 53
n = 65
n = 30
PAC
0.1
0.1 (75%)
<0.1 (18%)
<0.1 (8%)
0 (0%)
n=50
n = 16
n =7
n=0
CPL
34.0
12.9 (38%)
16.0 (47%)
3.5 (10%)
1.6 (5%)
n=135
n=252
n=66
n=38
GPL
5.4
1.5 (28%)
3.4 (63%)
0.1 (2%)
0.4 (7%)
n=97
n=182
n=8
n=14
NCE
18.5
10.5 (57%)
6.4 (35%)
1.1 (6%)
0.4(2%)
n=108
n=58
n=12
n=12
SAP
2.9
1.0 (34%)
1.4 (49%)
<0.1 (1%)
0.5 (16%)
n=75
n=112
n=5
n=33
TPL
7.6
3.1 (40%)
4.1 (53%)
0.1 (2%)
0.4 (5%)
n=106
n=186
n=12
n=19
WMT
4.7
1.6 (35%)
1.8 (38%)
0.5 (10%)
0.8 (17%)
n=148
n=134
=3
II
UJ
o
n=39
XER
5.4
2.1 (38%)
1.9 (36%)
0.4 (8%)
1.0 (19%)
n=100
n=131
n=31
n=47
1Numbers in table may not add to totals due to rounding.
2.6 Literature Cited
Cowardin LM, Carter V, Golet FC, LaRoe ET (1979) Classification of wetlands and deepwater habitats of
the United States. U. S. Department of the Interior, Fish and Wildlife Service, Washington, D.C.
Dahl TE (2006) Status and trends of wetlands in the conterminous United States 1998 to 2004. US
Department of Interior, Fish and Wildlife Service, Washington, DC
Dahl TE (2011) Status and trends of wetlands in the conterminous United States 2004 to 2009. US
Department of Interior, Fish and Wildlife Service, Washington, DC
Dahl TE, Bergeson MT (2009) Technical procedures for conducting status and trends of the Nation's
wetlands. US Fish and Wildlife Service, Division of Habitat and Resource Conservation, Washington, DC
Diaz-Ramos S, Stevens DL, Jr, Olsen AR (1996) EMAP Statistical Methods Manual. US Environmental
Protection Agency, Office of Research and Development, NHEERL-Western Ecology Division, Corvallis,
Oregon
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Dumelle, M, T. Kincaid, A.R. Olsen, and M. Weber. 2023. Spsurvey: Spatial Sampling Desing and Analysis in
R. Journal of Statistical Software, 105(3), 1-29. https://doi.orR/10.18637/iss.vl05.i03
Horvath EK, Christensen JR, Mehaffey MH, Neale AC (2017) Building a potential wetland restoration
indicator for the contiguous United States. Ecological indicators 83: 463-473.
Kincaid TM, Olsen AR (2019) spsurvey: Spatial Survey Design and Analysis. R package version 4.1.
Kloiber, SM (2010) Status and trends of wetlands in Minnesota: wetland quantity baseline. Minnesota
Department of Natural Resources, St Paul, Minnesota
Olsen AR, Kincaid TM, Kentula ME, Weber MH (2019) Survey design to assess condition of wetlands in the
United States. Environmental Monitoring and Assessment 191 (SI): 268, doi: 10.1007/sl0661-019-7322-
6.
Olsen AR, Kincaid TM, Payton Q (2012) Spatially balanced survey designs for natural resources. In: Gitzen
RA, Millspaugh JJ, Cooper AB, Licht DS (Eds.) PesiRn and Analysis of LonR-Term EcoloRical MonitorinR
Studies. Cambridge, UK, Cambridge University Press: 126-150.
Omernik JM (1987) Ecoregions of the Conterminous United States. Annals of the Association of American
Geographers 77: 118-125
R Core Team (2019) R: A language and environment for statistical computing. R Foundation for Statistical
Computing, Vienna, Austria. http://www.R-project.org
Stevens DL, Jr., Olsen AR (2003) Variance estimation for spatially balanced samples of environmental
resources. Environmetrics 14: 593-610
Stevens DL, Jr., Olsen AR (2004) Spatially-balanced sampling of natural resources. Journal of American
Statistical Association 99: 262-278
USEPA (2011) Level III Ecoregions of the Continental United States (revision of Omernik, 1987). US
Environmental Protection Agency, National Health and Environmental Effects Laboratory-Western
Ecology Division. Corvallis, Oregon
USEPA (2016a) National Wetland Condition Assessment 2016: Field Operations Manual. US
Environmental Protection Agency, Washington DC. EPA-843-R-15-007.
USEPA (2016b) National Wetland Condition Assessment 2016: Site Evaluation Guidelines. US
Environmental Protection Agency, Washington DC. EPA-843-R-15-010.
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual. US
Environmental Protection Agency, Washington DC. EPA-843-B-21-002.
USEPA (2021b) National Wetland Condition Assessment 2021: Site Evaluation Guidelines. US
Environmental Protection Agency, Washington DC. EPA-843-B-21-001.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
USFWS (2014) National Wetlands Inventory, US Department of the Interior, Fish and Wildlife Service,
Washington, D.C. https://www.fws.Rov/wetlands/Data/Metadata.html
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Chapter 3: Selection of Handpicked Sites
In addition to the probability sites identified by the survey design, handpicked sites are identified by
states, tribes, and other partners. These handpicked sites are suggested based on the expectation that
they are minimally disturbed and can be used as least-disturbed (or "reference") sites, although this is not
always the case (Herlihy 2008, 2019). The suggested handpicked sites were evaluated prior to the field
sampling using a screening process to eliminate those that are not likely to meet the criteria for the
NWCA.
3.1 Pre-Sampling Selection of Handpicked Sites
Candidate handpicked sites came from three sources:
1) Best Professional Judgment (BPJ) sites recommended by state, tribal, and federal entities with
responsibilities for wetlands;
2) Designated least-disturbed sites from other NARS with associated wetlands; and,
3) In-the-field replacements for sites from sources above that were determined not sampleable due
to access, permitting, or other constraints.
BPJ sites and least-disturbed sites designated from other NARS underwent the following screening
process.
3.1.1 Initial Screen
The initial screening step eliminated candidate handpicked sites not likely to meet the criteria for NWCA
sampling and to reduce the number of sites to a reasonable size for a manual evaluation employing
analysis of maps and aerial photos. Information provided by the person who suggested each site was
considered, and included wetland size and type, as well as data supporting whether a site was least
disturbed, e.g., scores from a Floristic Quality Assessment Index (FQAI) or Landscape Development Index
(LDI). Wetlands eliminated were typically small, rare types. In cases where many sites were submitted by
an entity, those ranking lower than others, given the data submitted, were eliminated from further
consideration.
3.1.2 Basic Screen
Candidate handpicked sites passing the initial screen were mapped in ArcGIS (exemplified in Figure 3-1).
Maps of each site with recent aerial imagery were assessed to determine if:
• The wetland at the site would support the establishment of a sampleable assessment area
o The wetland was in the target population for NWCA
o The wetland was equal or greater than 0.1 ha and at least 20-m wide
o Less than 10% of the area
¦ Contained water greater than 1-m deep,
¦ Had conditions that were unsafe or would make effective sampling impossible
(e.g., likely unstable substrate), and/or
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
¦ Was upland
o No hydrogeomorphic boundaries were crossed;
• The site was accessible with moderate effort; and,
• The site was greater than 1km away from a probability site.
If all these criteria were met, the sites were assessed for evidence of visual landscape disturbance.
NWCA 2016 Handpicked Ref Candidates
Site MNW11-185
State:
Latitude (DD): 47.705471
Longitude (DD): -94.223595
NWCA Wetland Type: PEM
Ecoregion: UMW
Ownership: Public
Name (if applicable):
Notes; Exceptional sites from 2011 MN intensification
Basic Screening Information
Sampleable AA can be established: ^ / N
Site is accessible with moderate effort: wf N
Not co-located with a probability site (<= 1 km): (£/ N
Visual Disturbance Information
N (none), Min (minimum), Mod+(mod or high)
Hydro figic modifications:
Agriculture or forestry:
Residents I, urban, or commercial:
Industrial -oif. gas, mining, etc:
Road networks:
(.1^ / Min / Mod+
(^y/ Min / Mod+
(U/Min / Mod+
Min / Mod+
none pved lo
(unjivS.pved hi
(toV Min / Mod+
$./ Min / Mod+
vN./ Min / Mod+
0/ Min / Mod+
none (guedjp^
Figure 3-1. Example of map created using ArcGIS software to evaluate candidate handpicked sites. Information
from an assessment of the aerial imagery was recorded for basic and landscape screening criteria.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
3.1.3 Landscape Screen
For candidate handpicked sites that passed the basic screen, aerial photos Figure 3-1 were used to
evaluate the presence of anthropogenic impact within buffers defined by 500m- and lkm-radius circles
centered on the likely location of the Assessment Area (AA) that would be used during field sampling.
First, the images were evaluated to determine the level of impact from the following types of
anthropogenic activities within the 500m- and lkm-radius buffer:
• Hydrologic modifications (e.g., linear features that would indicate the presence of ditches, dams,
or levees);
• Agricultural development (e.g., farm structures, row crops, horticultural fields, pastures) or
forestry activities (e.g., rows of trees, tree stumps and debris, logging roads, tree regeneration);
• Residential, urban, or commercial development (e.g., houses, retail malls, commercial buildings,
parking lots); and,
• Industrial development (oil and gas structures, mines, gravel pits, industrial facilities).
For each category of activity, the levels were noted as "none", "minimal" (the activity impacted less than
25% of the area), or "moderate and above" (the activity impacted more 25% or more of the area).
Next, the images were evaluated to determine the presence of road networks within the 500m- and lkm-
radius buffer. Road networks were categorized as "none", "unpaved only", "paved-low" (paved roads
impacted less than 25% of the area), or "paved-high" (paved roads impacted 25% or more of the area).
Sites with no impacts from anthropogenic activities and road networks were prioritized for sampling. Sites
with minimal impacts from anthropogenic activities and road networks ("unpaved", "paved-low") were
retained for potential use in regions with few non-disturbed candidate sites. Sites with "moderate" or
greater disturbance in the 500-m buffer were rejected outright.
3.1.4 Distribution of Handpicked Sites
Sites prioritized for sampling and retained for potential use were evaluated to assure, to the greatest
extent possible, adequate distribution across the regions and Wetland Groups likely to be used for
analysis and reporting. Site selection and distribution was also influenced by the availability of field crews
to sample handpicked sites in certain areas of the country. For example, EPA staffed regional field crews
were limited to sampling sites within their respective EPA Regions.
3.1.5 Replacement of Handpicked Sites Not Sampleable
At times, it was necessary to replace sites during the reconnaissance checks performed before sampling
or at the time of sampling. Sites were replaced during reconnaissance due to access issues, but also
because the Field Crew Leader acquired additional information that either 1) eliminated the site as a
candidate for use as "least-disturbed" (e.g., presence of invasive species) or 2) documented there was a
better, more appropriate candidate least-disturbed site. Sites were replaced at time of sampling primarily
due to access issues (e.g., too difficult to get to the exact location, last minute refusals by property
managers).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
3.1.6 Results
In the end, 123 handpicked sites (10 of which were sampled in 2011 and again (i.e., resampled) in 2016
and 2021) were selected through this screening process and sampled. Table 3-1 lists the final distribution
of the handpicked sites by the Five NWCA Aggregated Ecoregions and Wetland Group.
Table 3-1. Distribution of 123 handpicked sites sampled in 2021 by Five NWCA Aggregated Ecoregions and the
NWCA Wetland Group.
Five NWCA Aggregated Ecoregions
PRLH
PRLW
EH
EW
Total
Coastal Plains (CPL)
1
4
5
10
Eastern Mountains & Upper Midwest (EMU)
10
21
31
Interior Plains (IPL)
3
1
4
Western Valleys & Mountains (WMT)
28
16
4
48
Xeric West (XER)
24
5
1
30
Sum
66
47
10
123
NWCA Ecoregions
Coastal Plains (CPL)
Eastern Mountains and Upper Midwest (EMU)
HI Inland Plains (IPL)
HI Western Mountains (WMT)
Xeric (XER)
• 2021 Probability Sites
A 2021 Handpicked Sites
r.* - • •:
• * %* * 'i* vJ'A
v;
Figure 3-2. Map of the conterminous US showing distribution of handpicked sites (triangles) in relation to
probability sites (circles) sampled in the NWCA 2021.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
3.2 Literature Cited
Herlihy AT, Kentula ME, Magee TK, Lomnicky GA, Nahlik AM, Serenbetz G (2019) Strivingfor consistency
in the National Wetland Condition Assessment: developing a reference condition approach for assessing
wetlands at a continental scale. Environmental Monitoring and Assessment 191 (SI): 327, doi:
10.1007/s10661-019-7325-3
Herlihy AT, Paulsen SG, Van Sickle J, Stoddard JL, Hawkins CP, Yuan LL (2008) Striving for consistency in a
national assessment: the challenges of applying a reference-condition approach at a continental scale.
Journal of the North American Benthological Society 27: 860-877
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
National Wetland Condition Assessment: 2021 Technical Support Document
Dec 2024
-------
Chapter 4: Data Preparation
The tasks to produce the datasets used in the analysis are described in this chapter. The data checking
steps were designed to catch errors associated with missing, inconsistent, or illogical values. Other errors
were found and corrected during analysis using processes documented in subsequent chapters.
The master database for the NWCA 2021 includes:
1) Raw data collected by Field Crews and from laboratory processing of samples collected in the
field (USEPA 2021b, c).
2) Data documenting and characterizing the NWCA sites from the survey design.
3) Field and lab raw data, site information, and ancillary data combined for use in specific analyses.
4) Metrics calculated from raw data from the field forms and the laboratory results.
4.1 Data Entry and Review
4.1.1 Field Data
NWCA used standard field forms and centralized data management for data collected. Most field data
were collected electronically using an iPad with the NWCA field data mobile application. Following a
review for accuracy and completeness, field crews submitted the electronic forms directly from the
NWCA App to NARS IM, which automated upload to the NWCA 2021 SQL database. No paper field forms
were submitted in the 2021 survey.
4.1.1.1 Field Data Validation
Quality of field data were reviewed on a weekly, monthly and end of season basis using numerous
automated data quality checks. EPA staff and contractors then compiled a summary of data quality issues
which were sent to respective field crews to correct or provide additional comments. If field data could
not be corrected, crews were instructed to provide a comment as to why field data could not be collected
or measured. Corrected data and new comments were resubmitted from the NWCA App and updated in
the NARS IM NWCA 2024 SQL database.
4.1.2 Laboratory Data
Laboratory results were submitted to USEPA WRAPD staff, who checked the data for completeness and
obvious errors. Then the data files were transferred to NARS IM for incorporation into the master NWCA
database.
The water chemistry data produced by Consolidated Safety Services (CSS) located at PESD was handled by
a different process. CSS checks their results based on the approved Quality Assurance Project Plan and
the data files are transferred from CSS to the NARS IM through the Work Assignment Contract Officer
Representative (COR).
4.2 Quality Assurance Checks
There were three types of Quality Assurance (QA) checks completed before datasets were assembled for
analysis:
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
1) Verification of the fate of every sample point from the NWCA 2021 design;
2) Confirmation of longitudes and latitudes associated with the sites sampled; and
3) Data checks.
4.2.1 Verification of Points
Estimates of the wetland area falling into a particular condition category are based on the weight from
the survey design used to select the points to be sampled. In the NWCA survey design, the weight
indicates the wetland area in the NWCA target population represented by a point from the sample draw.
After the assessment is conducted, the weights are adjusted to account for additional sites (i.e., the
oversample points) evaluated when primary sites could not be sampled (e.g., due to denial of access,
being non-target).
All points in the design were reviewed to confirm which were sampled, and if not, the reasons why. Three
sources were used:
1) Information compiled during the desktop evaluation of sites (see the NWCA 2021 Site Evaluation
Guidelines (USEPA 2021c)), and documented by state and contractor field crews in spreadsheet
submissions to EPA during and after the 2021 field season,
2) Information recorded on Form PV-1 during a field evaluation performed prior to sampling (see
the NWCA 2021 Site Evaluation Guidelines (USEPA 2021c)), and
3) Information recorded on Form PV-1 at the time of sampling (see Chapter 3 in the NWCA 2021
Field Operations Manual (USEPA 2021a)).
Results from this evaluation were added to the database containing site information data from the NWCA
survey design and for the handpicked sites.
4.2.2 Confirmation of Coordinates Associated with the Sites Sampled
Longitudes and latitudes are taken at various key locations associated with field sampling (e.g., the
location of the point from the design). These coordinates are especially important if a point needs to be
relocated or shifted to accommodate sampling protocols (see Chapter 3 in the NWCA 2021 Field
Operations Manual (USEPA 2021a)). The coordinates are used to:
• Verify the relationship between the point coordinates from the design and those of the sampled
Assessment Area (AA) that represents the point;
• Tie the field data to landscape data from GIS layers; and
• Relocate the site and key locations of the field sampling protocol (e.g., the AA center, vegetation
plots) for resampling in future surveys.
Point coordinates from the design and the field were compared. The locations of points from the field
that were more than 60m from the corresponding design coordinates, i.e., that exceeded protocol
guideline {NWCA 2021 Site Evaluation Guidelines (USEPA 2021c)), were flagged.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
4.2.3 Data Checks
The first step in this series of checks was to assure all sites with data from a second field sampling (i.e.,
Visit 2, which is also known as the Quality Assurance Visit) had a corresponding initial sampling (i.e., Visit
1). Next, for all data types, computer code was written to generate a list of missing data, and checks were
performed to identify why they were missing (e.g., part of the sampling was not completed by the Field
Crew, data forms not submitted, etc.). Additional computer code was written to generate a list of data
not meeting a series of legal value and range tests. These tests were to confirm that:
• Data type was correct,
• Data fell within the valid range or legal value, and
• Units reported (especially for laboratory results) matched those expected.
Results of the checks were converted to Excel spreadsheets. Each potential error was evaluated by the IT
Team or the Indicator Lead using the original forms submitted by the Field Crew. A description of the
error and recommended resolution were recorded in the spreadsheet for each type of data and
incorporated into the master NWCA database. The Indicator Lead who would be the primary user of the
data was consulted in cases where the resolution of the issue could affect the results of the analysis.
4.3 Literature Cited
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual. US
Environmental Protection Agency, Washington DC. EPA-843-B-21-002.
USEPA (2021b) National Wetland Condition Assessment 2021: Laboratory Operations Manual. US
Environmental Protection Agency, Washington DC. EPA-843-B-21-003.
USEPA (2021c) National Wetland Condition Assessment 2021: Site Evaluation Guidelines. US
Environmental Protection Agency, Washington DC. EPA-843-B-21-001.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Chapter 5: Subpopulations
The conterminous United States is the broadest scale at which the NWCA results are reported. However,
the diversity in the Nation's landscape makes it important to assess aquatic resources in the appropriate
geographic setting. Regional variation in species composition, environmental conditions, and human-
caused disturbance often necessitates a finer scale, i.e., sub-national, to:
• Define quantitative criteria and thresholds for least-disturbed sites and most-disturbed sites;
• Define thresholds for categories of wetland condition and stressors, and
• Report wetland condition extent and stressor condition extent.
These tasks and the need for sub-national, geographic reporting units are inherent to all NARS
assessments.
USEPA's Environmental Monitoring and Assessment Program (EMAP) recommends as a general rule that,
absent information on the variability in the target population, 50 sites per subpopulation should be
assessed to increase the likelihood that the sample will be sufficient to make population estimates. For
example, the EPA Level III Ecoregions (Omernik 1987, USEPA 2011a) of the US were aggregated into nine
regions for the Wadeable Streams and National Lakes Assessments (USEPA 2006, 2009) to assure an
adequate number of sites per subpopulation. For the NWCA 2011, both regions and Wetland Groups
were used to report the results (USEPA 2016b). For the NWCA 2016, subpopulations for primary
reporting and for further investigations were developed for use in 2016 and subsequent surveys (Table
5-1). These subpopulation groups are discussed throughout the text of the Technical Support Document.
5.1 Literature Cited
Omernik JM (1987) Ecoregions of the conterminous United States. Annals of the Association of American
Geographers 77: 118-125
USEPA (2006) Wadeable Streams Assessment 2000-2004: A Collaborative Survey of the Nation's Streams.
EPA-841-B-06-002. US Environmental Protection Agency, Office of Water and Office of Research and
Development, Washington, DC
USEPA (2009) National Lakes Assessment 2007: A Collaborative Survey of the Nation's Lakes. EPA-841-R-
09-001. US Environmental Protection Agency, Office of Water and Office of Research and Development,
Washington, DC
USEPA (2011a) Level III Ecoregions of the Continental United States (revision of Omernik, 1987). US
Environmental Protection Agency, National Health and Environmental Effects Laboratory-Western
Ecology Division, Corvallis, OR
USEPA (2016b) National Wetland Condition Assessment 2011: A Collaborative Survey of the Nation's
Wetlands. US Environmental Protection Agency, Washington DC. EPA-843-R-15-005.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 5-1. Subpopulation information, including the parameter name that is used in the database, all the potential subpopulations included in each
subpopulation group, and a description of each subpopulation group.
Subpopulation Group
Parameter Name
Subpopulations
Description
Three Aggregated
Ecoregions
AG_EC03
EHIGH | PLNLOW | WMTNS
Codes for the aggregation of the Omernik Level III Ecoregions into three regions
(using 2015 boundaries); Eastern Highlands (EHIGH), Plains and Lowlands
(PLNLOW), Western Mountains (WMTNS).
Nine Aggregated
Ecoregions
AG_EC09
CPL | NAP | NPL | SAP | SPL | TPL |
UMW | WMT | XER
Codes for the aggregation of the Omernik Level III Ecoregions into nine regions
(using 2015 boundaries); Coastal Plains (CPL), Northern Appalachians (NAP),
Northern Plains (NPL), Southern Appalachians (SAP), Southern Plains (SPL),
Temperate Plains (TPL), Upper Midwest (UMW), Western Mountains (WMT), and
Xeric West (XER). For a visual, see the AG_EC09 tab in this workbook.
USFWSS&T Coastal
Regions
COAST_REG
Great Lakes Region | Gulf Coast
Region | North East Coast Region |
Not Coast | Pacific Coast Region |
South East Coast Region
US Fish and Wildlife Service Status and Trends Coastal Regions, including Great
Lakes Region, Gulf Coast Region, North East Coast Region, Pacific Coast Region,
and South East Coast Region. Sites that are not in a coastal region are designated
'Not Coast1.
EPA Regions
EPA_REG
Region_01 | Region_02 |
Region_03 | Region_04 |
Region_05 | Region_06 |
Region_07 | Region_08 |
Region_09 | Region_10
EPA Regions, responsible for the execution of programs within several states and
territories: Region 1 serving CT, ME, MA, NH, Rl, and VT, Region 2 serving NJ, NY,
Puerto Rico, and the US Virgin Islands, Region 3 serving DE, DC, MD, PA, VA, WV
and 7 federally recognized tribes, Region 4 serving AL, FL, GA, KY, MS, NC, SC, and
TN, Region 5 serving IL, IN, Ml, MN, OH, and Wl, Region 6 serving AR, LA, NM, OK,
and TX, Region 7 serving IA, KS, MO, and NE, Region 8 serving CO, MT, ND, SD, UT,
and WY, Region 9 serving AZ, CA, HI, NV, American Samoa, Commonwealth of the
Northern Mariana Islands, Federated States of Micronesia, Guam, Marshall
Islands, and Republic of Palau, and Region 10 serving AK, ID, OR, WA and 271
native tribes. For a visual, see the EPA_REGION tab in this workbook.
Federal Lands
FED_NONFED
FEDERAL | NON_FEDERAL
Using OWN_NARS, distinguishes Federal from Non-federal lands, with Federal
land comprised of 'BLM1, 'BOR1, 'DOD1, 'FWS1, 'NOAA1, 1NPS1, 'Other Fed', and
' US FS'.
Inland versusTidal
HYD_CLS
INLAND | TIDAL
Distinguishes tidal saline sites (HYD_CLS = TIDAL) from inland sites (HYD_CLS =
INLAND) combining information from the Aggregated S&T Class (WETCLS_GRP).
Specifically, EH + EW = TIDAL, and PRLH + PRLW = INLAND.
Major USGS Hydrologic
Basins
MAJ_BAS_NM
Arkansas-White-Red Region |
California Region | Great Basin
Region | Great Lakes Region |
Lower Colorado Region | Lower
Mississippi Region | Mid-Atlantic
Region | Missouri Region | New
England Region | Ohio-Tennessee
Region | Pacific Northwest Region |
Rio Grande-Texas-Gulf Region |
Souris-Red-Rainy Region | South-
Major US Geological Survey (USGS) hydrologic basins derived from NHD+ names,
with NHD+ codes in parenthesis: Arkansas-White-Red Region, California Region,
Great Basin Region, Great Lakes Region, Lower Colorado Region, Lower Mississippi
Region, Mid-Atlantic Region, Missouri Region, New England Region, Ohio-
Tennessee Region, Pacific Northwest Region, Rio Grande-Texas-Gulf Region,
Souris-Red-Rainy Region, South-Atlantic Region, Upper Colorado Region, Upper
Mississippi Region.
47
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Subpopulation Group
Parameter Name
Subpopulations
Description
Atlantic Region | Upper Colorado
Region | Upper Mississippi Region
Mississippi Basin
MISS_BASIN
MISSISS1 PPI_BAS 1N |
NOT_M ISSISSI PPI_BAS 1N
Designates whether a site is in the Mississippi Basin, which includes USGS
hydrologic basins (from MAJ_BAS_NM): Arkansas-White-Red Region, Lower
Mississippi Region, Missouri Region, Ohio-Tennessee Region, Upper Mississippi
Region.
USEPA National Estuary
Program
NEP_NAT
NEP | Not_NEP
Designates whether a site is in a USEPA National Estuary Program (NEP)
watershed. Does not include Chesapeake Bay.
Four NWCA Aggregated
Ecoregions
NWCA_EC04
CPL | EMU | IPL | W
Omernik Level III Ecoregions aggregated into Four NWCA Aggregated Ecoregions:
Coastal Plains (CPL), Eastern Mountains & Upper Midwest (EMU), Interior Plains
(IPL), and West (W). Note that inland and tidal saline sites are not distinguished.
For a visual, see the NWCA_EC04 tab in this workbook.
Five NWCA Aggregated
Ecoregions
NWCA_EC05
CPL | EMU | IPL | WMT | XER
Omernik Level III Ecoregions aggregated into Five NWCA Aggregated Ecoregions:
Coastal Plains (CPL), Eastern Mountains & Upper Midwest (EMU), Interior Plains
(IPL), Western Valleys & Mountains (WMT), and Xeric West (XER). Note that inland
and tidal saline sites are not distinguished. For a visual, see the NWCA_EC05 tab
in this workbook.
Four NWCA Aggregated
Ecoregions x Inland
versus Tidal
NWCA_EC04_HYD
CPL-INLAND | CPL-TIDAL | EMU-
INLAND | EMU-TIDAL | 1 PL-INLAND
| W-INLAND | W-TIDAL
Omernik Level III Ecoregions aggregated into Four NWCA Aggregated Ecoregions
(NWCA_EC04) and distinguished by inland sites (HYD_CLS = INLAND) or tidal
saline sites (HYD_CLS = TIDAL): Coastal Plains Inland (CPL-INLAND), Coastal Plains
Tidal (CPL-TIDAL), Eastern Mountains & Upper Midwest Inland (EMU-INLAND),
Eastern Mountains & Upper Midwest Tidal (EMU-TIDAL), Interior Plains Inland
(1 PL-INLAND), West Inland (W-INLAND), West Tidal (W-TIDAL). Note that there are
no Interior Plains Tidal sites, thus there is no 1PL-TI DAL subpopulation.
Five NWCA Aggregated
Ecoregions x Inland
versus Tidal
NWCA_EC05_HYD
CPL-INLAND | CPL-TIDAL | EMU-
INLAND | EMU-TIDAL | 1 PL-INLAND
| WMT-IN LAND | WMT-TIDAL |
XER-INLAND | XER-TIDAL
Omernik Level III Ecoregions aggregated into Five NWCA Aggregated Ecoregions
(NWCA_EC05) and distinguished by inland sites (HYD_CLS = INLAND) or tidal
saline sites (HYD_CLS = TIDAL): Coastal Plains Inland (CPL-INLAND), Coastal Plains
Tidal (CPL-TIDAL), Eastern Mountains & Upper Midwest Inland (EMU-INLAND),
Eastern Mountains & Upper Midwest Tidal (EMU-TIDAL), Interior Plains Inland
(1PL-INLAND), Western Valleys & Mountains Inland (WMT-INLAND), Western
Valleys & Mountains Tidal (WMT-TIDAL), Xeric West Inland (XER-INLAND), and
Xeric West Tidal (XER-TIDAL). Note that there are no Interior Plains Tidal sites,
thus there is no 1 PL-TI DAL subpopulation.
Land Ownership
OWN_NARS
BLM | BOR | City | County | DOD |
FWS | NGO | NOAA | Non-Federal |
NPS | Other Fed | Regional | State |
Tribal | USFS
Designates land ownership: Bureau of Land Management (BLM), Bureau of
Reclamation (BOR), City, County, Department of Defense (DOD), Fish and Wildlife
Survey (FWS), Non Governmental Organizations (NGO), National Oceanic and
Atmospheric Administration (NOAA), National Park Service (NPS), other federal
(Other Fed), Regional, State, Tribal, and US Forest Service (USFS) lands. Non
Federal lands are designated 'Non-Federal1.
States
PSTL_CODE
AL | AR | AZ | CA | CO | CT | DE |
FL | GA | IA | ID | IL | IN | KS | KY |
US State: Alabama (AL), Arizona (AZ), Arkansas (AR), California (CA), Colorado
(CO), Connecticut (CT), Delaware (DE), Florida (FL), Georgia (GA), Idaho (ID),
48
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Subpopulation Group
Parameter Name
Subpopulations
Description
LA | MA | MD | ME | Ml | MN |
MO | MS | MT | NC | ND | NE | NH
| NJ | NM | NV | NY | OH | OK | OR
| PA | Rl | SC | SD | TN | TX | UT |
VA | VT | WA | Wl | WV | WY
Illinois (IL), Indiana (IN), Iowa (IA), Kansas (KS), Kentucky (KY), Louisiana (LA),
Maine (ME), Maryland (MD), Massachusetts (MA), Michigan (Ml), Minnesota
(MN), Mississippi (MS), Missouri (MO), Montana (MT), Nebraska (NE), Nevada
(NV), New Hampshire (NH), New Jersey (NJ), New Mexico (NM), New York (NY),
North Carolina (NC), North Dakota (ND), Ohio (OH), Oklahoma (OK), Oregon (OR),
Pennsylvania (PA), Rhode Island (Rl), South Carolina (SC), South Dakota (SD),
Tennessee (TN), Texas (TX), Utah (UT), Vermont (VT), Virginia (VA), Washington
(WA), West Virginia (WV), Wisconsin (Wl), Wyoming (WY)
Ten NWCA Reporting
Groups
RPTGRP_10
ALL-EH | ALL-EW | CPL-PRLH | CPL-
PRLW | EMU-PRLH | EMU-PRLW |
IPL-PRLH | IPL-PRLW | W-PRLH | W-
PRLW
Ten NWCA Reporting Groups used for the NWCA analysis that combines Four
NWCA Aggregated Ecoregions (NWCA_EC04) and Aggregated S&T Classes
(NWCA_WET_GRP): All Estuarine Herbaceous (ALL-EH), All Estuarine Woody (ALL-
EW), Coastal Plains Palustrine, Riverine, and Lacustrine Herbaceous (CPL-PRLH),
Coastal Plains Palustrine, Riverine, and Lacustrine Woody (CPL-PRLW), Eastern
Mountains & Upper Midwest Palustrine, Riverine, and Lacustrine Herbaceous
(EMU-PRLH), Eastern Mountains & Upper Midwest Palustrine, Riverine, and
Lacustrine Woody (EMU-PRLW), Interior Plains Palustrine, Riverine, and Lacustrine
Herbaceous (IPL-PRLH), Interior Plains Palustrine, Riverine, and Lacustrine Woody
(IPL-PRLW), West Palustrine, Riverine, and Lacustrine Herbaceous (W-PRLH), West
Palustrine, Riverine, and Lacustrine Woody (W-PRLW). Note that estuarine sites
(ALL-EH and ALL-EW) are combined for the contiguous US.
Twelve NWCA Reporting
Groups
RPTGRP_12
ALL-EH | ALL-EW | CPL-PRLH | CPL-
PRLW | EMU-PRLH | EMU-PRLW |
IPL-PRLH | IPL-PRLW | WMT-PRLH |
WMT-PRLW | XER-PRLH | XER-
PRLW
Twelve NWCA Reporting Groups used for the NWCA analysis that combines Five
NWCA Aggregated Ecoregions (NWCA_EC05) and Aggregated S&T Classes
(NWCA_WET_GRP): All Estuarine Herbaceous (ALL-EH), All Estuarine Woody (ALL-
EW), Coastal Plains Palustrine, Riverine, and Lacustrine Herbaceous (CPL-PRLH),
Coastal Plains Palustrine, Riverine, and Lacustrine Woody (CPL-PRLW), Eastern
Mountains & Upper Midwest Palustrine, Riverine, and Lacustrine Herbaceous
(EMU-PRLH), Eastern Mountains & Upper Midwest Palustrine, Riverine, and
Lacustrine Woody (EMU-PRLW), Interior Plains Palustrine, Riverine, and Lacustrine
Herbaceous (IPL-PRLH), Interior Plains Palustrine, Riverine, and Lacustrine Woody
(IPL-PRLW), Western Valleys & Mountains Palustrine, Riverine, and Lacustrine
Herbaceous (WMT-PRLH), Western Valleys & Mountains Palustrine, Riverine, and
Lacustrine Woody (WMT-PRLW), Xeric West Palustrine, Riverine, and Lacustrine
Herbaceous (XER-PRLH), Xeric West Palustrine, Riverine, and Lacustrine Woody
(XER-PRLW). Note that estuarine sites (ALL-EH and ALL-EW) are combined for the
contiguous US.
Ten Reporting Units
RPTJJNIT
ARW | ATL | GFC | GPL | ICP | NCE
| PAC | SAP | TPL | WVM
Ten Reporting Units created using a combination of information from AG_EC09
and WETCLS_GRP to distinguish regions of inland sites from regions of tidal saline
sites: Atlantic Coast (ATL), Arid West (ARW), Gulf & Florida Coasts (GFC), Great
Plains (GPL), Inland Coastal Plains (ICP), North Central East (NCE), Pacific Coast
(PAC), Southern Appalachians (SAP), Temperate Plains (TPL), and Western Valleys
& Mountains (WVM). Note that the inland region names have been changed to
49
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Subpopulation Group
Parameter Name
Subpopulations
Description
distinguish regions with similar boundaries but combine inland and tidal saline site
(e.g., CPLfrom NWCA_EC04and NWCA_EC05 includes ATL, IPL, and GPL sites
from RPTJJNIT). For a visual, see the RPTJJNIT tab in this workbook. The DATA
CROSSWALK tab in this workbook explains how the units were derived (and how
they refer to other parameters).
Five Reporting Units
RPT_UNIT_5
EMU | ICP | PLN | TDL | WST
Five Reporting Units for reporting that uses tidal saline wetlands as a distinct
region from the Four Aggregated NWCA Ecoregions (NWCA_EC04), and created
by collapsing the Ten Reporting Units (RPTJJNIT): Eastern Mountains & Upper
Midwest (EMU = NCE + SAP), Inland Coastal Plains (ICP), Plains (PLN = GPL + TPL),
Tidal Saline (TDL = ATL + GFC + PAC), West (WST = ARW + WVM).
Six Reporting Units
RPT_UNIT_6
ARW | EMU | ICP | PLN | TDL |
WVM
Six Reporting Units for reporting that uses tidal saline wetlands as a distinct region
from the Five Aggregated NWCA Ecoregions (NWCA_EC05), and created by
collapsing the Ten Reporting Units (RPT_UNIT): Arid West (ARW), Eastern
Mountains & Upper Midwest (EMU = NCE + SAP), Inland Coastal Plains (ICP),
Plains (PLN = GPL + TPL), Tidal Saline (TDL = ATL + GFC + PAC), Western Valleys &
Mountains (WVM).
12-Ecoregion x Wetland
Group Reporting Units
RPTJJNIT12
ARW-H | ARW-W | EMU-H | EMU-
W | ICP-H | ICP-W | PLN-H | PLN-W
| TDL-H | TDL-W | WVM-H | WVM-
W
Twelve reporting units derived from the combination of Six Reporting Units
(RPT_UNIT_6) and Wetland Groups (WETCLS_GRP): Arid West Herbaceous (ARW-
H), Arid West Woody (ARW-W), Eastern Mountains & Upper Midwest Herbaceous
(EMU-H), Eastern Mountains & Upper Midwest Woody (EMU-W), Inland Coastal
Plains Herbaceous (ICP-H), Inland Coastal Plains Woody (ICP-W), Plains
Herbaceous (PLN-H), Plains Woody (PLN-W), Tidal Saline Herbaceous (TDL-H),
Tidal Saline Woody (TDL-W), Western Valleys & Mountains Herbaceous (WVM-H),
Western Valleys & Mountains Woody (WVM-W).
Twenty Reporting Units
RPTJJNIT20
ARW-H | ARW-W | ATL-H | ATL-W |
GFC-H | GFC-W | GPL-H | GPL-W |
ICP-H | ICP-W | NCE-H | NCE-W |
PAC-H | PAC-W | SAP-H | SAP-W |
TPL-H | TPL-W | WVM-H | WVM-W
Twenty Reporting Units created using a combination of information from
AG_EC09 and WETCLS_GRP to distinguish regions of inland sites from regions of
tidal saline sites, and WETCLS_GRP to distinguish herbaceous (H) from woody (W)
dominated sites: Atlantic Coast Herbaceous (ATL-H), Atlantic Coast Woody (ATL-
W), Arid West Herbaceous (ARW-H), Arid West Woody (ARW-W), Gulf & Florida
Coasts Herbaceous (GFC-H), Gulf & Florida Coasts Woody (GFC-W), Great Plains
Herbaceous (GPL-H), Great Plains Woody (GPL-W), Inland Coastal Plains
Herbaceous (ICP-H), Inland Coastal Plains Woody (ICP-W), North Central East
Herbaceous (NCE-H), North Central East Woody (NCE-W), Pacific Coast
Herbaceous (PAC-H), Pacific Coast Woody (PAC-W), Southern Appalachians
Herbaceous (SAP-H), Southern Appalachians Woody (SAP-W), Temperate Plains
Herbaceous (TPL-H), Temperate Plains Woody (TPL-W), Western Valleys &
Mountains Herbaceous (WVM-H), and Western Valleys & Mountains Woody
(WVM-W). Note that the inland region names have been changed to distinguish
regions with similar boundaries but combine inland and tidal saline site (e.g., CPL
from NWCA_EC04 and NWCA_EC05 includes ATL, IPL, and GPL sites from
RPT_UNIT). For a visual, see the RPTJJNIT tab in this workbook. The DATA
50
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Subpopulation Group
Parameter Name
Subpopulations
Description
CROSSWALK tab in this workbook explains how the units were derived (and how
they refer to other parameters).
USFWSS&T Wetland
Classes
WETCLS_EVAL
E2EM | E2SS | NONE | PEM | PF |
PFO | PSS | PUBPAB
US Fish and Wildlife Service Status and Trends wetland class designated in the
field on Form AA-2 on date of sampling. If evaluated in field but not sampled, then
wetland class is assigned from field visit. If site only evaluated in office, then
wetland class assigned at that time. If no site evaluation information on wetland
class, then wetland class assigned wetland class used for the survey design. Hand-
picked sites should be assigned during field sampling. Wetland classes use FWS
S&T classes: Estuarine Intertidal Emergent (E2EM), Estuarine Intertidal
Forest/Shrub (E2SS), Palustrine Emergent (PEM), Palustrine Farmed (PF),
Palustrine Forested (PFO), Palustrine Shrub (PSS), and Palustrine Unconsolidated
Bottom/Aquatic Bed (PUBPAB). See Reference Card AA-3, Side A in the 2011 and
2016 NWCA Field Operations Manuals for details. NONE only applies to non-
sampled sites.
Wetland Groups
WETCLS_GRP
EH | EW | PRLH | PRLW
Aggregated US Fish and Wildlife Service Status and Trends wetland class based on
the design that combines wetland type and dominant vegetation type for
reporting: Estuarine Herbaceous (EH), Estuarine Woody (EW), Palustrine, Riverine,
and Lacustrine Herbaceous (PRLH), and Palustrine, Riverine, and Lacustrine
Woody (PRLW).
Hydrogeomorphically-
Altered
WETCLS_ALT
HGM_ALTERED |
HGM_NOT_ALTERED
Using WETCLS_HGM2, distinguishes HGM altered sites from not altered sites
using QAed and validated values (HGM_CLASS_VALID and HGM_SUBCLASS_VALID
in tbIASSESSMENT) designated in the field on Form AA-2 on date of sampling.
HGM_ALTERED includes 'DEPRESSION_ALT', 'FLATS_ALT', 'LACUSTRINE_ALT',
1RIVERINE_ALT', 'SLOPE_ALT', and 'TIDAL_ALT' while HGM_NOT_ALTERED includes
'DEPRESSION1, 'FLATS', 'LACUSTRINE', 'RIVERINE', 'SLOPE', and 'TIDAL'.
Hydrogeomorphic Classes
WETCLS_HGM
DEPRESSION | FLATS | LACUSTRINE
| RIVERINE | SLOPE | TIDAL
Hydrogeomorphic (HGM) class from QAed and validated values
(HGM_CLASS_VALID in tbIASSESSMENT) designated in the field on Form AA-2 on
date of sampling, including depression (DEPRESSION), flats (FLATS), lacustrine
fringe (LACUSTRINE), riverine (RIVERINE), slope (SLOPE), and tidal (TIDAL) wetland
classes. See Reference Card AA-3, Side B in the 2011 and 2016 NWCA Field
Operation Manuals for details.
Hydrogeomorphic Classes
Distinguishing Natural
versus Altered
WETCLS_HGM2
DEPRESSION | DEPRESSION_ALT |
FLATS | FLATS_ALT | LACUSTRINE |
LACUSTRINE_ALT | RIVERINE |
RIVERINE_ALT | SLOPE | SLOPE_ALT
| TIDAL | TIDAL_ALT
Hydrogeomorphic (HGM) classes, separated into natural and altered HGM
subclasses from QAed and validated values (HGM_CLASS_VALID and
HGM_SUBCLASS_VALID in tbIASSESSMENT) designated in the field on Form AA-2
on date of sampling. Unaltered HGM classes include depression (DEPRESSION),
flats (FLATS), lacustrine fringe (LACUSTRINE), riverine (RIVERINE), slope (SLOPE),
and tidal (TIDAL). Altered HGM subclasses are indicated by an appended '_ALT'
and include DEPRESSION_ALT (includes subclasses 'Closed, Human Impounded',
'Closed, Human Excavated', 'Closed, Human Excavated and Impounded', 'Open,
Human Impounded', 'Open, Human Excavated', 'Open, Human Excavated and
Impounded', FLATS_ALT (includes subclass 'Human Altered'), LACUSTRINE_ALT
51
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Subpopulation Group
Parameter Name
Subpopulations
Description
(includes subclass 'Artificially Flooded1), RIVERINE_ALT (includes subclass 'Human
Altered'), SLOPE_ALT (includes subclass 'Human Altered'), and TIDAL_ALT
(includes subclass 'Human Altered'). See Reference Card AA-3, Side B in the 2011
and 2016 NWCA Field Operations Manuals for details. Note that '_ALT' indicates
the historic HGM class that should be at the site, denoting that it has been altered
(e.g., a site that was historically FLATS but excavated into a depression would be
designated as FLATS_ALT).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
52
-------
Chapter 6: Assigning Disturbance Class
Anthropogenic disturbances to wetlands vary in impacts and intensities across different regions of the
United States (USEPA 2016a,b, Lomnicky et al. 2019). Following the practice of previous NARS
assessments (e.g., USEPA 2006, 2008, 2009, 2016a), the NWCA uses a quantitative definition of
disturbance using physical, chemical, and biological data collected at wetland sites sampled as part of the
NWCA. These data reflect a continuous gradient of anthropogenic disturbance - ranging from no
observable or measurable anthropogenic impacts to highly altered wetland sites. Wetland sites that fall
along this continuous disturbance gradient are assigned to one of three disturbance classes: "least
disturbed", "intermediate disturbed", or "most disturbed" (Figure 6-1, USEPA 2016a). Thus, thresholds
that delineate the boundaries of each disturbance class must be set.
, 1 y 1 v
,SX
Disturbance Gradient
Reference Sites
Figure 6-1. Diagram of the disturbance gradient used in the NWCA with three classes of disturbance.
Because pristine conditions are uncommon or absent in most places, the NWCA uses the characteristics
found in least-disturbed sites as "reference". Least-disturbed sites are those with the best available
physical, chemical, and biological condition given the current status of the landscape in which the site is
located (Stoddard et al. 2006). Least-disturbed status for the NWCA is defined using a set of explicit
quantitative criteria for specific disturbance indicators. It is expected that these least-disturbed sites will
represent good ecological condition (Karr 1991, Dale and Beyeler 2001, Stoddard et al. 2006, 2008),
although this may not always be the case given that "least disturbed" in some areas of the country still
has considerable disturbance.
The planning for the NWCA assumes:
• The survey design provides a representative sample of the target population;
• Least-disturbed sites reflect the functional capacity and delivery of services typical of a given
wetland type in a particular landscape setting (e.g., ecoregion, watershed); and,
• Thresholds developed from data collected on-site and used to define disturbance classes provide
benchmarks against which to compare assessment results.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Least- and most-disturbed sites are needed in the development of condition indicators - both for the
evaluation of candidate metrics (Chapter 8:) that may reflect ecological condition and for the
development of Vegetation Multi metric Indices (VMMIs) (Chapter 9:). Specifically, least-disturbed sites
are used in setting thresholds for good, fair, and poor condition based on VMMI values (Magee et al.
2019a, Herlihy et al. 2019).
This chapter documents the complex process for 1) developing quantitative definitions of site-level
anthropogenic disturbance based on physical, chemical, and biological data, 2) establishing least- and
most-disturbed thresholds, and 3) assigning sampled sites to least-, intermediate-, and most-disturbed
classes. The process for calculating indices and metrics and for assigning disturbance class is summarized
in An Illustrative Guide to Assigning Disturbance Class in Six Steps found in Section 6.9, Appendix A.
Due to delays in the availability of soil heavy metal results from the 2021 survey, the NWCA analysis team
did not have complete data to assign disturbance classes to sites sampled in 2021. Analysis of the 2021
data was done using the set of reference sites established for the 2011 and 2016 surveys, described in
this chapter. When the soil data are available, the NWCA analysis team will assign disturbance classes to
2021 sites and evaluate whether the additional data warrants updating assessment benchmarks for
certain indicators.
6.1 Sites Used to Establish the Disturbance Gradient
Data from a total of 1,987 unique probability and handpicked sites across both the NWCA 2011 and the
NWCA 2016 were used in a screening process to establish a disturbance gradient (Table 6-1). The
sampling events at these 1,987 unique sites are referred to as Index Visits, as they include only the first
sampling visit (i.e., Visit 1) and only the 2016 site data (i.e., not the 2011 site data) if a site from 2011 was
also sampled in 2016. In other words, if the same site was sampled in both 2011 and 2016, the most
recent Visit 1 was used as the Index Visit. For the 208 resampiedsites, we chose to use the 2016 data
over the 2011 data because of improvements in the field protocols for collecting disturbance information,
and because using data associated with the most recent survey is standard across other NARS. The
probability sites were either from the NWCA design or a related probability design produced by NARS for
a state intensification (Chapter 2:). The handpicked sites included those identified for and sampled in the
2016 survey (Chapter 3:) and the handpicked sites sampled in the 2011 survey (USEPA 2016a).
Table 6-1. The number of Visit 1 (VI) probability and handpicked sites sampled in 2011 and 2016, with their totals.
Additionally, the numbers of resampied sites are reported in paratheses to indicate that these are subtracted from
the subtotals above. The total number of unique probability and handpicked sites are reported with the final
number of Index Visit sites (in the red cell) used in the establishment of the NWCA disturbance gradient. Note that
this table does not include the 96 Visit 2 sites sampled in 2011 and 94 Visit 2 sites sampled in 2016, which are only
used to calculate Signal-to-Noise ratios for some indicators/metrics (see Chapter 8: for details).
VI PROBABILITY
HANDPICKED
SURVEY YEAR
(n-sites)
(n-sites)
TOTAL
2011 NWCA
967
171
1138
2016 NWCA
967
90
1057
SUBTOTAL
1934
261
2195
2011 Sites Resampied in 2016
(207)
(1)
(208)
TOTAL UNIQUE SITES
1727
260
1987
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
6.2 Establishing a Disturbance Gradient
The general steps in the process of establishing a disturbance gradient are:
• Develop indices or metrics that reflect anthropogenic disturbance,
• Set thresholds for "least disturbed" for each index or metric,
• Set thresholds for "most disturbed" for each index or metric, and
• Use a screening process to define each site as "least", "intermediate", or "most disturbed"
(Herlihy etal. 2008, 2019).
To develop the disturbance gradient for the NWCA, a stepwise process was used in which sites were first
screened using physical indices, then by chemical indices, and finally through a biological metric. Methods
for calculating the indices and metrics used in screening are explicitly discussed in the NWCA 2016
Technical Support Document (Chapter 11, 12). The general process for setting thresholds and assigning
disturbance classes are described in the following sections.
6.2.1 Indices and Metrics
Physical, chemical, and biological data collected in the field and laboratory were evaluated for use in
screening sites to establish the disturbance gradient. Indices and metricswere chosen based on evidence
of a strong association with anthropogenic stress and on the robustness of the data. The indices and
metrics used are described in Table 6-2.
Table 6-2. Indices and metrics used in NWCA 2016 to establish the disturbance gradient. Final indices and metrics
for which thresholds were created are in uppercase, bold type.
Screen Type Data Type Indices and Metrics Reference
Physical
Human-
Mediated
Physical
Alterations
• Vegetation Removal (PALT_VEGRMV)
• Vegetation Replacement (PALT_VEGREP)
• Water Addition/Subtraction (PALT_WADSUB)
• Water Obstruction (PALT_WOBSTR)
• Soil Hardening (PALT_SOHARD)
• Surface Modification (PALT_SOMODF)
• PALT_ANY
• PALT_SUM
2016 TSD,
Chapter 11
Chemical
Soil
Chemistry
• Enrichment Factor (EF)}* EF_MAX
• Heavy Metal Index (HMI)
2016 TSD,
Chapter 12
Biological
Vegetation
• Relative Percent Cover of Nonnative (alien and cryptogenic)
Plant Species (XRCOV_AC)
Section 6.6
Physical and chemical indices were used to define least- and most-disturbed sites based primarily on
abiotic characteristics under the variable name REF_NWCA_ABIOTIC. The biological metric was used to
further screen the least-disturbed sites designated in REF_NWCA_ABIOTIC, resulting in some of these
sites being rejected from least-disturbed status. The resulting final disturbance class designations are
found under the variable name REF_NWCA.
Although water chemistry is a part of the NWCA field protocol, only 56% and 65% of the wetlands in 2011
and 2016, respectively, sampled across both Visit 1 and Visit 2 had sufficient surface water to collect and
analyze. In addition, wetland hydroperiod- especially during the growing season when NWCA sampling
occurred - can greatly influence water chemistry (e.g., nutrients can become highly concentrated during
drawdowns) and introduce bias into the types of wetlands sampled for water chemistry (see Kentula et al.
2020). Thus, water chemistry was excluded from the generation of the disturbance gradient. However,
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
the water chemistry analyses, including how disturbance classes were assigned to just the wetland sites
sampled for water chemistry, are presented in a stand-alone chapter of this report (Chapter 13:).
Additionally, while we were able to gather landscape data (e.g., land use within a 1-km buffer of the AA)
using GIS layers, we opted not to use these data to screen sites for the disturbance gradient. This was for
two reasons: 1) the GIS layers are less precise than the data we were able to gather in the field, and 2) it
is possible that wetlands in good condition exist in what is considered an "impacted" landscape.
Therefore, we used only information directly measured by Field Crews on the ground to establish the
disturbance gradient.
6.2.2 Setting Least-Disturbed Thresholds
For each of the indices and metrics in Table 6-2, a least-disturbed threshold was set. Physical and
chemical thresholds were set independently by the subpopulation group Five Reporting Units
(RPT_UNIT_5), as the extent of human disturbance can vary greatly among regions. Following the
definition of least-disturbed as the best-available sites (Stoddard et al. 2006), thresholds for "least
disturbed" in ubiquitously impacted regions may be greater than those for "intermediate disturbed" or
even "most disturbed" in regions that have greater amounts of intact area. Initially, physical and chemical
thresholds were set to zero human disturbance in all regions. However, if a subpopulation (i.e., region)
did not have a sufficient number of least-disturbed sites with these stringent thresholds, the thresholds
were relaxed so that approximately 15-25% of the sites in the subpopulation passed the screens to obtain
a sufficient number of least-disturbed sites for data analysis. The set of least-disturbed sites identified
using the physical and chemical screens were further screened using a biological metric, and any sites
that exceeded 10% relative cover of nonnative plants were rejected from least-disturbed status.
6.2.3 Setting Most-Disturbed Thresholds
Most-disturbed sites were defined using a screening process in the same manner as for least-disturbed
sites. The same physical and chemical measures of disturbance were used, and thresholds for most
disturbed were set for each measure. If any single threshold for any measure was exceeded, the site was
considered a most-disturbed site. As "most disturbed" is a relative definition, our objective was to define
approximately 20-30% of the sites in a subpopulation as "most disturbed", and thresholds were set
accordingly.
6.2.4 Classifying Disturbance at Each Site for each Sampling Visit
Finally, disturbance status was assigned to each site for each of its sampling visits (i.e., Visit 1, Visit 2, and
both 2011 and 2016 visits for resampled sites). Disturbance status was assigned by screening each site
visit to test for exceedance of least- and most-disturbed thresholds. Sites were first screened using the
physical and chemical indices and metrics. If a site exceeded the most-disturbed thresholds, it was
considered most-disturbed. If any single physical or chemical threshold was exceeded at a site, it was not
considered "least-disturbed". Sites identified as least-disturbed based on this screen were further
screened using the biological metric. Thus, the final set of least-disturbed sites were those that were
below the thresholds for all physical, chemical, and biological measures. Sites not falling into either least-
or most-disturbed categories were classified as having intermediate disturbance.
The following sections provide details about the data used to develop thresholds for each index or metric
in Table 6-2 and the thresholds used for defining least- and most-disturbed sites.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
6.3 Human-Mediated Physical Alteration Screens and Thresholds
Human-Mediated Physical Alteration scores were calculated for each NWCA site sampled in 2011 and
2016 using methods summarized in Section 6.9, Appendix A: Steps 1 and 2. Thresholds were developed
for Five Reporting Units (RPT_UNIT_5, see Table 5-1 in Chapter 5:), which include the subpopulations
Tidal Saline (TDL), Inland Coastal Plains (ICP), Eastern Mountains & Upper Midwest (EMU), Plains (PLN),
and West (WST). Two screens that integrate scores from all six physical alteration indices (VEGRMV,
VEGRPL, WADSUB, WOBSTR, SOHARD, and SOMODF, see Table 6-2) were applied to each site using the
thresholds described in Table 6-3:
• PALT_ANY - For any given site, the PALT_ANY screen for "least disturbed" was applied by
considering each of the six physical alteration indices individually. If the score for any one index
(i.e., the maximum score among all six indices) was greater than a threshold, the site was no
longer considered "least disturbed". The threshold varies by subpopulation, ranging from 0 to <
20, meaning that a least-disturbed site may have (up to) a few observed physical alterations in
the buffer plots, but no observations of physical alterations in the AA.
• PALT_SUM - The PALT_SUM screen for "least disturbed" was developed to capture instances
where there were multiple observed physical alterations at a site, but those observances were
spread across multiple indices and, therefore, may have passed the PALT_ANY screen despite
moderate to high levels of overall disturbance. For any given site, the PALT_SUM screen was
applied by considering the sum of the scores from all six physical alteration indices. If the sum of
scores for all six indices was greater than a threshold, the site was no longer considered "least
disturbed". Like PALT_ANY, the threshold varies by subpopulation, ranging from 0 to < 40,
meaning that there were no or few observations of physical alterations regardless of index in the
AA or buffer.
Sites may pass the PALT_ANY screen and fail the PALT_SUM screen if there are several observations in
buffer plots within different physical alteration categories. Sites ultimately classified as "least disturbed"
had to pass both the PALT_ANY and the PALT_SUM screens (in addition to other chemical and biological
screens described in the following sections of this Technical Support Document). The least-disturbed
thresholds and the number of sites that passed the physical alteration screens (and were considered
candidate least-disturbed sites) are presented in Table 6-3a.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 6-3. a) Least-disturbed thresholds and b) most-disturbed thresholds for the two physical alteration screens
and the number of sites that passed the screens (i.e., are considered candidate "least disturbed" or "most
disturbed") presented for Five Reporting Units (RPT_UNIT_5).
gj Physical
Screens for
Least-Disturbed
Sites
Tidal Saline
(TDL)
Inland Coastal
Plains (ICP)
Eastern
Mountains &
Upper Midwest
(EMU)
Plains (PLN)
West (WST)
PALT ANY
0
0
0
10
20
PALT SUM
0
0
0
10
40
n-sites
200
100
117
100
83
U\ Physical
' Screens for
Most-Disturbed
Sites
Tidal Saline
(TDL)
Inland Coastal
Plains (ICP)
Eastern
Mountains &
Upper Midwest
(EMU)
Plains (PLN)
West (WST)
PALT ANY
30
50
40
50
70
PALT SUM
60
100
80
100
140
n-sites
87
95
61
84
87
The most-disturbed sites on the disturbance gradient were defined using a screening process in the same
manner as for least-disturbed sites. Thresholds for "most disturbed" were set for PALT_ANY and
PALT_SUM. If any single threshold for any measure was exceeded, the site was considered a most-
disturbed site. As "most disturbed" is a relative definition, our objective was to classify approximately 20-
30% of the sites in a subpopulation as "most disturbed", and thresholds were set accordingly. The most-
disturbed thresholds and the number of sites considered candidate "most disturbed" are presented in
Table 6-3b.
Sites that did not meet "least disturbed" or "most disturbed" threshold criteria were classified as
"intermediate disturbed".
For some sites, data were not collected at all on the H-l Form and/or B-l Form, or an insufficient number
of buffer plots (<5) were sampled. In these cases, the sites could not be evaluated using the physical
screens (i.e., PALT_ANY and PALT_SUM) and were categorized as "unknown" (coded as "?") for their
physical screen disturbance class.
6.4 Chemical Screens and Thresholds
Two chemical screens were used as the second set of screens (with the first set being the physical screens
discussed in the previous section) to assign abiotic disturbance class (REF_NWCA_ABIOTIC) to each site.
These screens are 1) the Heavy Metal Index (HMI) and 2) the Maximum Enrichment Factor (EF_MAX), the
calculations for which are summarized in Section 6.9, Appendix A: Steps 3 - 5. In brief, the Enrichment
Factor (EF) is calculated for each of 12 heavy metals at each site to capture the degree to which soils are
enriched. Using the EF information, the HMI is calculated, which indicates the number of heavy metals
with moderate enrichment or greater (EF > 3). Finally, the EF_MAX is calculated, indicating the highest
degree to which a site was contaminated by any of the 12 heavy metals.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
58
-------
Sites ultimately assigned as abiotic "least disturbed" had to pass the PALT_ANY and the PALT_SUM screens
and the HMI and EF_MAX screens. National thresholds for "least disturbed" were used for both the HMI
and EF_MAX and were:
• HMI < 1
• EF_MAX < 5
In other words, regardless of the region in which a site was located, for a site to be considered "least
disturbed", only one heavy metal EF could be equal to or above three, and the EF of any heavy metal had
to be less than five. Although national thresholds were used for the HMI and EF_MAX screens, region-
specific heavy metal background concentrations were used in the EF calculation (specifically, as the
denominator) (see Section 6.9, Appendix A: Step 4 for details). The chemical screen thresholds for "least
disturbed" and the number of sites that passed the chemical screens (i.e., were considered abiotic "least
disturbed" (REF_NWCA_ABIOTIC)) are presented in Table 6-4a.
Table 6-4. a) Least-disturbed thresholds and b) most-disturbed thresholds for the two chemical screens and the
number of sites that passed the screens (i.e., are considered abiotic "least disturbed" or "most disturbed")
presented for Five Reporting Units (RPT_UNIT_5).
Chemical
' Screens for
Least-Disturbed
Sites
Tidal Saline
(TDL)
Inland Coastal
Plains (ICP)
Eastern
Mountains &
Upper Midwest
(EMU)
Plains (PLN)
West (WST)
HMI
< 1
< 1
< 1
< 1
< 1
EF MAX
<5
<5
< 5
< 5
<5
n-sites
180
96
105
98
68
Chemical
' Screens for
Most-Disturbed
Sites
Tidal Saline
(TDL)
Inland Coastal
Plains (ICP)
Eastern
Mountains &
Upper Midwest
(EMU)
Plains (PLN)
West (WST)
HMI
> 1
> 1
> 1
> 1
> 1
EF MAX
> 10
> 10
> 10
> 10
> 10
n-sites
109
105
72
88
101
The most-disturbed sites on the disturbance gradient were defined using a screening process in the same
manner as for least-disturbed sites. National thresholds for most disturbed were set for the HMI and
EF_MAX and were:
• HMI > 1
• EF_MAX > 10
If any threshold of either chemical screen (i.e., HMI or EF_MAX) was exceeded, the site was considered
"most-disturbed". In other words, regardless of the region in which a site was located, any more than one
heavy metal EF equal to or above three or an EF of any heavy metal greater than five resulted in a site
regarded as "most disturbed". As "most disturbed" is a relative definition, our objective was to define
approximately 20-30% of the sites in a subpopulation as most disturbed, and thresholds were set
accordingly. In particular, the EF_MAX was set above ten to equate a level of severe enrichment with a
most-disturbed site. The most-disturbed thresholds and the number of sites considered "abiotic most
disturbed" are presented in Table 6-4b.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
It is important to note that the thresholds established for heavy metals do not reflect toxicity thresholds.
These thresholds are indicators of human disturbance.
For some sites, soil chemistry samples were not collected. In these cases, the sites could not be evaluated
using the chemical screens (i.e., HMI and EF_MAX) and were categorized as "unknown" for their chemical
screen disturbance class.
6.5 Abiotic Disturbance Class Assignments
Physical and chemical screens were combined to assign sites to abiotic disturbance classes of "least
disturbed", "intermediate disturbed", "most disturbed", and "unknown", coded in the data as
REF_NWCA_ABIOTIC. In general, the highest disturbance class between the physical and chemical screens
is used to assign the abiotic disturbance class. If physical alteration data were missing from a site, the
abiotic disturbance class was assigned as "unknown". If soil chemistry data were missing from a site, the
abiotic disturbance class was set to that of the physical screen disturbance class1. The application of rules
used to assign abiotic disturbance classes is illustrated in Figure 6-2.
For any single site:
Physical Screen
Disturbance Class
Chemical Screen
Disturbance Class
Abiotic
Disturbance Class
Figure 6-2. A visual summary of how rules for assigning abiotic disturbance classes based on the physical and
chemical screens are applied to a site, where L = "least disturbed", I = "intermediate disturbed", M = "most
disturbed", and ? = "unknown". Note that the physical and chemical screens were evaluated together to
determine the abiotic disturbance class assignment for a site.
A summary of the number of sites within each abiotic disturbance class are reported by region
(RPT_UNIT_5) in Table 6-5.
xThe decision to use the physical screen disturbance level instead of assigning "unknown" when soil chemistry data
were missing from a site was based on the low prevalence of sites with "intermediate disturbance" or "high
disturbance" assignments based on the chemical screens alone.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 6-5. n-sites of abiotic disturbance class assignments (REF_NWCA_ABIOTIC) reported by region (RPT_UNIT_5)
for Visit 1, Index Visit 2011 and 2016 sites
Least
Intermediate
Most
Regional
Region
Disturbed (L)
Disturbed (1)
Disturbed (M)
Unknown (?)
Totals
Tidal Saline (TDL)
180
170
109
3
462
Inland Coastal Plains (ICP)
96
207
105
4
412
E. Mts & Upper Midwest (EMU)
105
172
72
1
350
Plains (PLN)
98
163
88
2
351
West (WST)
68
242
101
1
412
National Totals
547
954
475
11
1987
6.6 Biological Screen and Threshold
Many sites designated as "least disturbed" using the physical and chemical screens had high relative
cover of nonnative plants, and such sites do not reflect natural vegetation conditions (Sala et al. 1996,
Lesica 1997, Vitousek et al. 1997, Ehrenfeld 2003, Dukes and Mooney 2004, Magee et al. 2010, 2019b).
Consequently, the set of abiotic least-disturbed sites (REF_NWCA_ABIOTIC == L) were screened with a
biological screen, resulting in a new set of final least-disturbed sites.
The biological screen was comprised of a single metric - the relative percent cover of nonnative (alien
and cryptogenic) plants species (XRCOV_AC), summarized in Section 6.9, Appendix A: Step 6. Relative
percent cover of nonnative plant species (XRCOV_AC) is calculated as the relative cover of alien and
cryptogenic species across the five sampled 100-m2 vegetation plots2 as a percentage of total plant cover,
or:
a cover of all alien + cryptogenic taxa across 5 plots \
— - - 1 *100
cover of all individual taxa across 5 plots #
The final set of least-disturbed sites for the NWCA (see the REF_NWCA variable) had to pass the
PALT_ANY, PALT_SUM, HMI, and EF_MAX least-disturbed screens and the XRCOV_AC least-disturbed
screen. The national threshold used for "least disturbed" was:
• XRCOV AC < 10%
In other words, regardless of region, for a site to be considered "least disturbed", nonnative plants had to
make up less than 10% of the total vegetation cover. The biological screen threshold and the number of
sites that passed this screen (i.e., assigned "least-disturbed" status) are presented in Table 6-6.
2 Data describing the abundance (percent cover) of all vascular species were collected in five 100-m2 vegetation
plots systematically distributed within each NWCA Assessment Area according to the Vegetation Protocol (USEPA
2011b, USEPA 2016c). Data collection methods are summarized in Section 7.3. In addition, each individual plant
taxon-state pair identified in NWCA 2011 and 2016 was assigned to a native status category: native, introduced,
adventive, cryptogenic, or unknown (see Chapter 7:, Section 7.8 and Table 7-5).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 6-6. The least-disturbed threshold for the biological screen, and the number of sites passing the screen (and
thus, are assigned final "least-disturbed" status as indicated in REF_NWCA) for the Five Reporting Units
(RPT_UNIT_5).
Biological
Screen for
Least-Disturbed
Sites
Tidal Saline
(TDL)
Inland Coastal
Plains (ICP)
Eastern
Mountains &
Upper Midwest
(EMU)
Plains (PLN)
West (WST)
XRCOV AC
< 10%
< 10%
< 10%
< 10%
< 10%
n-sites
149
86
101
53
50
Contrary to the methods used for physical and chemical disturbance gradient screens, the biological
screen was not used to designate most-disturbed sites. Instead, abiotic least-disturbed sites (based on the
physical and chemical screens) that were rejected using the biological screen were reassigned as
intermediate-disturbed sites. Thus, the set of least- and intermediate-disturbed sites are different for
REF_NWCA_ABIOTIC and REF_NWCA. However, the set of most-disturbed sites are the same.
6.7 Final Disturbance Class Assignments
The final disturbance class site assignments, which include "least disturbed", "intermediate disturbed",
and "most disturbed", are recorded as the variable, REF_NWCA, and was used for evaluation of
vegetation candidate metrics and for VMMI development based on data from NWCA 2011 and 2016 (see
Chapter 8: and Chapter 9:).
A summary of final disturbance designations (REF_NWCA) reporting the number of sites within each
disturbance class by region (RPT_UNIT_5) is provided in Table 6-7 and mapped in Figure 6-3.
Table 6-7. n-sites within final disturbance class assignments (REF_NWCA) reported by region (RPT_UNIT_5) for
Visit 1, Index Visit 2011 and 2016 sites. Note that two sites (one from TDL and another from ICP) were dropped
due to insufficient vegetation data and assigned as "unknown".
Least
Intermediate
Most
Regional
Region
Disturbed (L)
Disturbed (1)
Disturbed (M)
Unknown (?)
Totals
Tidal Saline (TDL)
149
201
108
4
462
Inland Coastal Plains (ICP)
86
216
105
5
412
E. Mts & Upper Midwest (EMU)
101
176
72
1
350
Plains (PLN)
53
208
88
2
351
West (WST)
50
260
101
1
412
National Totals
439
1061
474
13
1987
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
62
-------
a 2011 Least-Disturbed Sites
2011 Intermediate-Disturbed Sites
~ 2011 Most-Disturbed Sites
• 2016 Least-Disturbed Sites
2016 Intermediate-Disturbed Sites
• 2016 Most-Disturbed Sites
Figure 6-3, Map of sampled sites and their final disturbance class (REF_NWCA) assignments.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
6.8 Literature Cited
Dale VH, Beyeler SC (2001) Challenges in the development and use of ecological indicators. Ecological
Indicators 1: 3-10.
Dukes JS, Mooney HA (2004) Disruption of ecosystem processes in western North America by invasive
species. Revista Chilena de Historia Natural 77: 411-437.
Ehrenfeld JG (2003) Effects of exotic plant invasions on soil nutrient cycling processes. Ecosystems 6: 503-
523.
Herlihy AT, Paulsen SG, Van Sickle J, Stoddard JL, Hawkins CP, Yuan LL (2008) Striving for consistency in a
national assessment: the challenges of applying a reference-condition approach at a continental scale.
Journal of the North American Benthological Society 27: 860-877
Herlihy AT, Paulsen SG, Kentula ME, Magee TE, Nahlik AM, Lomnicky GA (2019) Assessingthe relative and
attributable risk of stressors to wetland condition across the conterminous United States. Environmental
Monitoring and Assessment 191 (SI): 320. DOI: 10.1007/sl0661-019-7313-7.
Karr JR (1991) Biological integrity: A long-neglected aspect of water resource management. Ecological
Applications 1: 66-84.
Kentula ME, Nahlik AM, Paulsen SG, Magee TK (2020) Wetland assessment: Beyond the traditional water
quality perspective. In: Summers JK (Ed.) Water Quality: Science, Assessments and Policy. IntechOpen,
DOI: 10.5772/intechopen.92583.
Lesica P (1997) Spread of Phalaris arundinacea adversely impacts the endangered plant Howellia aquatilis.
Great Basin Naturalist 57: 366-368.
Lomnicky GA, Herlihy AT, Kaufmann PR (2019) Quantifying the extent of human disturbance activities and
anthropogenic stressors in wetlands across the conterminous United States - results from the National
Wetland Condition Assessment. Environmental Monitoring and Assessment 191 (SI): 324, doi:
10.1007/s 10661-019-7314-6.
Magee TK, Blocksom KA, Fennessy MS (2019a) A national-scale vegetation multimetric index (VMMI) as
an indicator of wetland condition across the conterminous United States. Environmental Monitoring and
Assessment 191 (SI): 322, doi: 10.1007/sl0661-019-7324-4.
Magee TK, Blocksom KA, Herlihy AT, &. Nahlik AM (2019b) Characterizing nonnative plants in wetlands
across the conterminous United States. Environmental Monitoring and Assessment 191 (SI): 344, doi:
10.1007/s 10661-019-7317-3. https://link.springer.com/article/10.1007/sl0661-019-7317-3
Magee TK, Ringold PL, Bollman MA, Ernst T (2010) Index of Alien Impact: a method for evaluating
potential ecological impact of alien plant species. Environmental Management 45: 759-778.
Sala A, Smith SD, Devitt DA (1996) Water use by Tamarix ramosissima and associated phreatophytes.
Ecological Applications 6: 888-898.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Stoddard JL, Larsen DP, Hawkins CP, Johnson PK, Norris RH (2006) Setting expectations for the ecological
condition of streams: the concept of reference condition. Ecological Applications 16: 1267-1276.
Stoddard JL, Herlihy AT, Peck DV, Hughes RM, Whittier TR, Tarquinio E (2008) A process for creating
multimetric indices for large-scale aquatic surveys. Journal of the North American Bethological Society 27:
878-891.
USEPA (2006) Wadeable Streams Assessment: A Collaborative Survey of the Nation's Streams. US
Environmental Protection Agency, Office of Water and Office of Research and Development, Washington,
DC.
USEPA (2008) Ecological Research Program Multi-Year Plan FY2008-2014: February 2008 Review Draft. US
Environmental Protection Agency, Office of Research and Development, Washington, DC.
USEPA (2009) National Lakes Assessment 2007: A Collaborative Survey of the Nation's Lakes. EPA-841-R-
09-001. US Environmental Protection Agency, Office of Water and Office of Research and Development,
Washington, DC.
USEPA (2016a) National Wetland Condition Assessment 2011 Technical Report. EPA-843-R-15-006. US
Environmental Protection Agency, Washington DC.
USEPA (2016b) National Wetland Condition Assessment 2011: A Collaborative Survey of the Nation's
Wetlands. EPA-843-R-15-005. US Environmental Protection Agency, Washington DC.
USEPA (2023) National Wetland Condition Assessment 2016: Technical Support Document. EPA-841-B-
23-001. US Environmental Protection Agency, Washington, DC.
Vitousek PM, D'Antonio CM, Loope LL, Rejmanek M, Westbrooks R (1997) Introduced species: a
significant component of human-caused global change. New Zealand Journal of Ecology 86: 33212-33218.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
6.9 Appendix A: Illustrative Guide to Assigning Disturbance Class in Six Steps
CALCULATE PHYSICAL ALTERATION INDICES
For each Visit 1, Index Visit probability and handpicked site, calculate a score for each of six Physical Alteration indices (VEGRMV,
VEGRPL, WADSUB, WOBSTR, SOHARD, and SOMODF). Evaluate the highest score among all six Physical Alteration indices to
determine PALT_ANY and the total score among all six Physical Aleration indices to determine PALT_SUM.
For each of the six Physical Alteration indices (VEGRMV, VEGRPL, WADSUB,
WOBSTR, SOHARD, SOMODF), sum the score of the observed metrics (i.e., eight
metrics in each of the six indices) in the AA and buffer plots.
Where, VEGRMV
VEGRPL
WADSUB
WOBSTR
SOHARD
SOMODF
Vegetation Removal
Vegetation Replacement
Water Addition/Subtraction
Water Obstruction
Soil Hardening
Surface Modification
1
Calculate PALT ANY and PALT SUM.
Where, PALT_ANY=
The maximum score among all six
Physical Alteration indices.
Where, PALT_SUM=
The total score by summing the scores of
all six Physical Alteration indices.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
66
-------
SCREEN SITES THROUGH PHYSICAL ALTERATION INDICES
Screen sites using least-disturbed and most-disturbed Physical Alteration thresholds in five regions using Physical Alteration indices:
PALT_ANY (VEGRMV, VEGRPL, WADSUB, WOBSTR, SOHARD, SOMODF) and PALT_SUM. All Visit 1 sites (probability and
handpicked) sampled in 2011 and 2016 are evaluated. Sites that pass the screens remain candidates for least- or most-disturbed sites.
PHYSICAL ALTERATION THRESHOLDS
LEflST-DISf IIBBIB I ICR TBI I MOST-DISTURBED I ICR I TBI
PALT_ANY 20 10 0 0 0 PALT_ANY 70 50 | 40 50 | 30
PALT_SUM 40 i 0 0 i 0 PALT_SUM 140 : 100 i 80 100 : 60
n-sites 83 | 100 117 i 100 200 n-sites 87 84 G1 95 87
67
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
ESTIMATE SOIL HEAVT METAL BACKGROUNDS
Using the candidate least-disturbed sites that passed the Physical Alteration screens in Steps 1 and 2, calculate the 75th percentile of
the concentration of each of 12 heavy metals (Ag, Cd, Co, Cr, Cu, Ni, Pb, Sb, Sn, V, W, Zn) within each of Five Reporting Units
(RPT_UNIT_5). Note: heavy metal concentration graphs illustrate how 75th percentiles are determined and do not show actual results.
westtwsTi
1 1
Analysis of
Heavy Metal Background
Site Distribution
n+7 |
n+6
n+5_
O
O
jS n+3
Percentiles:
95th
r?5tfr-
T
50th
25th
5th
CANDIDATE LEAST- / MOST-DISTURBED SITES
L.
Piains (PLN)
Analysis of
Heavy Metal Background
Site Distribution
n+7
n+6
n+5
O
O
2 n+3
o
2
^ n+2
1
Percentiles:
95th
-rWth
I
50th
Analysis of
Heavy Metal Background
Site Distribution
n+7 |
Percentiles:
O
o
2 n+3
0)
5
I
95th
Mfe
50th
25th
5th
n+7
o
(0
i_
n+5
c
0)
c
n+4
o
o
co
n+3
0
>.
>
n+2
0)
T
n+'l
0
Inland Coastal
Plains (ICP)
i
Analysis of
Heavy Metal Background
Site Distribution
Percentiles:
I
75th
50th
T
25th
5th
L.
Tidal Saline
(TDL)
nr
Analysis of
Heavy Metal Background
Site Distribution
n+7 ¦
n+6
n+5
O
O
2 n+3
>
o
X
Percentiles:
I
,75th
1
25th
HEAVY METAL BACKGROUND CONCENTDATIONS (ppm)
75th Percentile
(ppm)
Ag
Cd
Co
Cr
Cu
Ni
Pb
Sb
Sn
v
w
Zn
WST
0.19
0.46
8.99
39.7
28.5
22.6
24.3
0.47
1.46
65.4
0.19
81.7
PLN
0.17
0.55
9.17
38.8
19.5
23.3
26.4
0.34
1.45
65.6
0.04
97.2
EMU
0.15
0.82
5.17
22.9
15.2
13.8
37.4
0.40
1.41
33.9
0.18
61.7
ICP
0.09
0.26
8.06
39.4
14.2
18.3
24.6
0.31
1.47
52.9
0.05
64.6
TDL
0.15
0.15
7.30
538
17.2
21 4
25.1
0.29
1.69
75.8
0.06
73.0
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATE ENRICHMENT FACTORS ANO HEAVY METAL INDEX
For each site, calculate an Enrichment Factor, or EF, (Chen et al. 2007) for each of the 12 heavy metals using the regional heavy metal
background estimated from Step 3. Then, combine the 12 EFs into a Heavy Metal Index (HMI) and calculate the maximum
EF (EF_MAX) for each site, which are used in combination as the chemical screen in abiotic reference site selection (Step 5).
Enrichment Factor = EF
For each site, EF_AG
EF_CD
EF_CO
EF_CR
EF_CU
EF_NI
EF_PB
EF_SB
EF_SN
EF_V
EF_W
EF ZN
-(
Observed heavy metal concentration at a site
Regional 75th percentile heavy metal background
)
Where, EF < 1
no enrichment
EF < 3
minor enrichment
EF = 3-5
moderate enrichment
EF = 5-10
moderately severe enrichment
EF = 10-25
severe enrichment
EF = 25-50
very severe enrichment
EF > 50
extremely severe enrichment
^number of heavy metals with EF > 3 = Heavy Metal Index = HMI
Where, the maximum of the HMI can be 12 for any
site (i.e., if all 12 heavy metal EFs are
equal to or greater than 3, indicating
moderate enrichment or greater depending
on the EF values).
Maximum Enrichment Factor = EF_MAX = maximum value of the 12 heavy metal EFs
Chen, C.W., C.M. Kao, C.F. Chen, & C.D. Dong (2007) Distribution and accumulation of heavy metals in the sediments of Kaohsiung Harbor, Taiwan. Chemosphere 66(8): 1431-1440.
National Wetland Condition Assessment: 2021 Technical Support Document
-------
SCREEN SITES THROUGH HEAVY METAL INDICES
Using the sites that passed the Physical Alteration screens (i.e., candidate least-disturbed / most-disturbed sites), rescreen the sites using
the heavy metal screens (HMI and EF_MAX) in Five Reporting Units (RPT_UNIT_5). Sites that pass the screens are the final abiotic
least- or most-disturbed sites (i.e., sites passed all the physical and chemical screens).
U
&
westmvsT)
Plains (PIN! Eastern Mountains inland Coastal
a Upper Midwest (EMII] Plains (ICP)
Tidal Saline
(TDD
>
I
u
w
HEAVY METAL THRESHOLDS
LEAST-DISTURBED
WST
ICP
TDI
HMI
< 1
< 1
< 1
< 1
< 1
EF_MAX
< 5
< 5
<5
< 5
< 5
n-sites
68
98
105
96
180
MOST-DISTURRED
WST j
ICP
TDI
HMI
> 1
> 1
> 1
> 1
> 1
EF_MAX
> 10
> 10
> 10
> 10
> 10
n-sites
101
88
72
105
109
70
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
u
s
u
«S
to
0.
yd
M
CALCULATE RELATIVE PERCENT COVER OF NONNATIVES & SCREEN
Calculate the relative percent cover of normative (alien and cryptogenic) plant species, XRCOV_AC, of each site. Then, screen only the
abiotic least-disturbed sites (i.e., sites that passed all the physical and chemical screens) using a national biological screen with a threshold
of XRCOV_AC < 10%. Any site that does not pass the biological screen is classified as "intermediate disturbed".
For each site, calculate the relative percent cover of normative (alien and cryptogenic)
plant species (XRCOV AC):
XRCOV AC =
(I
cover of all alien + cryptogenic taxa across 5 plots
cover of all individual taxa across 5 plots
)
100
C3
C/D
CD
&
bhi
H—
C/D
*
WestlWSTl
4 A
O
Eastern Mountains Inland Coastal Tidal Saline
Plains IICP1 (TDD
x—r: ^
M
(A
t/i
o£
i
LJ
NATIONAL THRESHOLD:
XRCOV AC <10%
Sites that pass the biological screen are designated as final least-disturbed, or "reference", sites.
Sites that do not pass are classified as "intermediate-disturbed".
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
71
-------
National Wetland Condition Assessment: 2021 Technical Support Document
Dec 2024
-------
Chapter 7: Vegetation Analysis Overview, Data Acquisition, and
Preparation
7.1 Background
The status of natural vegetation has been
increasingly and effectively used as an
indicator of ecological condition in wetlands
(Mack and Ken tula 2010, USEPA 2016a,
Magee et al. 2019a). In wetland ecosystems,
vegetation provides biodiversity, primary
productivity, habitat for organisms in other
trophic levels, and contributes to energy,
nutrient, and sediment or soil dynamics
(Mitsch and Gosselink 2007, Tiner 1999).
Wetland vegetation both responds to and
influences hydrology, water chemistry, soils,
and other components of the biophysical
habitat of wetlands. Because plants respond
directly to physical, chemical, and biological
conditions at multiple temporal and spatial
scales, they can be excellent indicators of
ecological condition or stress (Mclntyre and
Lavorel 1994, Mclntyre et al. 1999). For
example, wetland plant species 1) represent
diverse adaptations, ecological tolerances,
and life history strategies, and 2) integrate
environmental conditions, species
interactions, and human-caused disturbance. As a result, many human-mediated disturbances are
reflected in shifts in the presence or abundance of particular plant species, plant functional groups
(Quetier et al. 2007), plant communities (Galatowitsch et al. 1999, DeKeyser et al. 2003), and vegetation
structural elements (Mack 2007). In addition, some vegetation metrics are likely to be more prominently
expressed in particular wetland types, and some wetland types may be more likely than others be
subjected to higher anthropogenic disturbance levels or to be less resilient to this disturbance (USEPA
2016b, Magee et al 2019a).
Data describing plant species composition (species identity, presence, and abundance) and vegetation
structure were collected in the 2011, 2016, and 2021 NWCA surveys. Such data are powerful, robust,
relatively easy to gather and can be summarized into myriad candidate metrics that may be related to
ecological condition (USEPA 2002, Mack and Kentula 2010, Magee et al. 2019a). In addition to reflecting
ecological condition, some plant species groups can be indicators of stress to wetlands. Nonnative plant
species, in particular, are recognized as indicators of declining ecological condition, or as stressors to
ecological condition (Magee et al. 2008, Ringold 2008, Magee et al. 2010, Magee et al. 2019b).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
73
-------
Vegetation Multimetric Indices (VMMI) and a Nonnative Plant Indicator of Stress (NNPI) were used to
evaluate wetland condition based on the NWCA 2021 and changes and trends in condition observed
across all NWCA surveys.
Vegetation Multimetric Indices (VMMI) of Condition
Background,/VMMIs include several metrics describing different aspects of the observed vegetation that
together can reflect wetland condition in relation to least-disturbed wetland sites. In developing VMMIs,
individual candidate vegetation metrics are evaluated for their utility in distinguishing least disturbed sites
from those that are most disturbed. Several of most effective metrics are then selected and combined
into a VMMI as an indicator of wetland condition. VMMIs commonly include a suite of vegetation metrics
(representing aspects of plant communities, vegetation structure, and functional or life history guilds)
(e.g., DeKeyser et al. 2003, Miller et al. 2006, Reiss 2006, Rocchio 2007, Veselka et al. 2010, Euliss and
Mushet 2011, Genet 2012, Rooney et al. 2012, Deimeke et al. 2013, Wilson et al. 2013).
NWCA VMMIs-.
• NWCA 2011\ A four-metric VMMI that is applicable across the national-scale of the conterminous
US was developed and employed in assessing wetland condition based on data from the 2011
NWCA (USEPA 2016a, USEPA 2016b, Magee et al. 2019a).
• NWCA 2016, NWCA 2021\ For the NWCA 2016 analysis, the combined number of wetland sites
sampled in the 2011 and 2016 NWCAs provided sufficient data to allow development of separate
VMMIs for four major Wetland Groups: Estuarine Herbaceous, Estuarine Woody, Inland
Herbaceous, and Inland Woody. These four VMMIs were used to assess wetland condition for
NWCA 2021.
Nonnative Plant Indicator of Stress (NNPI)
The NNPI was first developed for the NWCA 2011 (USEPA 2016a, USEPA 2016b, Magee et al. 2019b), and
has been used for subsequent surveys in 2016 and 2021. The NNPI incorporates attributes of richness,
occurrence, and abundance for nonnative (alien and cryptogenic) plant species and was used to assess
the extent of potential stress to wetlands from nonnative plants (Chapter 10). In addition to describing
stress to a wetland, the NNPI can also be viewed as an indicator of vegetation condition.
7.2 Overview of Vegetation Analysis Process
As the primary biotic indicator of wetland condition for the NWCA, vegetation is a major component of
the NWCA analysis pathway (Figure 1-1). Evaluating vegetation in the NWCA included three sequential
phases, each with several major analysis steps (Figure 7-1). The first phase, data acquisition and
preparation, is covered in this chapter. The second phase, describing the prerequisite steps for vegetation
indicator development, including candidate metric calculation and evaluation is covered in Chapter 8. The
third phase, describing condition and stress, is covered in Chapter 9, which details the development of
the NWCA VMMIs used in 2016 and 2021 analyses, and Chapter 10, which summarizes the Nonnative
Plant Indicator.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Acquire and
Validate Field &
Lab Data
Standardize
Plant Taxonomy
Acquire/Develop
Species Trait
Characteristics
Identify Least
and Most
Disturbed Sites
Evaluate Plant
Species
Composition
Calculate and
Screen Candidate
Metrics
3. Describing Condition and Stress
r
Develop Vegetation
Multimetric Indices
(VMMIs)
Estimate Wetland
Area in Good, Fair,
Poor Condition
based on VMMIs
Calculate Nonnative
Plant Indicator (NNPI)
Estimate Wetland Area
in Good, Fair, Poor, or
Very Poor Condition in
Relation to Stress from
Nonnative Plants
Figure 7-1. Overview of vegetation data preparation and analysis steps used in assessing NWCA wetlands.
The three analysis elements depicted in Figure 7-1, their included steps, and the Sections or Chapters in
which they are discussed are listed below:
1. Data Acquisition and Preparation
• Collect field data (Section 7.3)
• Validate raw data (Section 7.4)
• Standardize plant species taxonomy (Section 7.5)
• Acquire or develop plant species trait information used in development of candidate vegetation
metrics (Sections 7.6 - 7.9)
2. Steps Prerequisite to indica tor Development
• Define disturbance gradients by identifying least- and most-disturbed sampled sites (Section 8.2
and Chapter 6)
• Evaluate plant species composition in relation to ecoregion and wetland type to maximize
homogeneity within groups of sites for analysis and potential VMMI development (Section 8.3)
• Use raw vegetation data (Section 7.11 Appendix B) and species trait information (Sections 7.6 -
7.9) to calculate candidate vegetation metrics (Section 8.4)
• Evaluate candidate vegetation metrics for potential utility for use in VMMI development (Section
8.5).
3. Description of Ecological Condition and Stress
• For each of four major Wetland Groups, develop a Vegetation Multimetric Index that reflects
wetland condition along an anthropogenic disturbance gradient (Chapter 9).
75
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
• Describe how VMMIs are used to estimate wetland area in good, fair, and poor condition across
the conterminous US and within various wetland subpopulations (Chapter 15:),
• Calculate Nonnative Plant Indicator of stress (NNPI) (Chapter 10)
• Describe how NNPI is used to estimate wetland area, across the conterminous US and within
various wetland subpopulations, that has good, fair, poor, or very poor condition in relation to
stress from nonnative plants (Chapter 15:).
7.3 Vegetation Data Collection
The Vegetation Protocols for the NWCA were designed to address the survey objectives, while meeting
logistics constraints of completion in one sampling day per site by a four-person Field Crew. The sampling
protocols are detailed in the NWCA 2021 Field Operations Manual (FOM) (USEPA 2021a), which has
updates and additions compared to the 2011 and 2016 FOMs (USEPA 2011a, USEPA 2016c). A brief
overview of the standardized NWCA field sampling and plant data collection protocols, and identification
protocols for unknown plants represented by collected specimens, is provided in the following two
subsections.
7.3.1 Field Sampling
To facilitate consistency and quality in vegetation data collection, Field Crews were provided with:
• Standardized training in vegetation sampling protocols prior to beginning sampling; and
• An Assistance Visit from NWCA experts to a sample site to answer any crew questions about
protocol implementation, generally during the first week of field sampling.
Vegetation data for the NWCA were collected during the peak growing season when most plants were in
flower or fruit to optimize species identification and characterization of species abundance. At each
NWCA sample point location, data were gathered in five 100-m2 Vegetation (Veg) Plots.
• The five Veg Plots were placed systematically in a >2 hectare Assessment Area (AA) at each site.
• In each plot vegetation data were collected across the entire 100-m2 plot and also in smaller
nested quadrats within each plot.
• Alternate configurations for AA shape and plot locations were used only, when necessary, as
determined by rules related to specific site conditions (USEPA 2021a).
• Standard AA and Veg Plot layouts are illustrated in Figure 7-2, the configuration of each plot is
shown in Figure 7-3.
• Key activities of the vegetation sampling protocol, and the data collected in each step are
provided in the flowchart in Figure 7-4.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Figure 7-2. Standard NWCA Assessment Area (AA) (shaded circular area) and standard layout of Vegetation Plots.
100-m2 Veg Plot
10m
3.16m
1.00m
1m2
10m2
/
NE Quadrat Nest
10m2
1m2
\
SW Quadrat Nest
Figure 7-3. Diagram of a Vegetation Plot illustrating plot boundaries and positions of nested quadrats.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
77
-------
Figure 7-4. Overview of vegetation data collection protocol for the NWCA 2021 (USEPA 2021a).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
78
-------
7.3.2 Identification of Unknown Plant Species
Plant species observed in the Veg Plots at each site that could not be identified in the field, were collected
for later identification. Specimen collection, labeling, specimen preservation (pressing and drying),
shipping or delivering dried specimens to a designated laboratory or herbarium, and specimen tracking
were completed according to standard protocols described in the NWCA 2021 Field Operations Manual
(USEPA 2021a).
Identification of unknown plant taxa was guided by protocols in the NWCA 2021 Laboratory Operations
Manual (USEPA 2021b). Unknown plant specimens from each Field Crew were identified at a specific
designated regional laboratory or herbarium (hereafter, lab) by one or more lab botanists. As quality
control for the identification process, ten percent of the lab identifications for unknowns were
independently verified by another botanist at the lab. Lab botanists maintained a detailed spreadsheet
that included for each unknown specimen collected in the field: the collection number and pseudonym
from the field collection, the location of collection (plot and site number), date of sampling, the name
assigned during lab identification based on a regional flora, and any notes related to the identification.
The lab botanists also reviewed quality assurance (QA) plant voucher specimens (5 per site) collected by
the field crew to confirm identifications of these species by the field botanists and to calculate percent
taxonomic agreement between lab and field botanists.
All identifications of unknown and QA vouchers were recorded in the NWCA 2021 Plant ID Lab
Spreadsheets (an Excel workbook). This Excel workbook includes instructions for required information to
be recorded for each specimen, as well as user information tabs that provide quick reference lists and
instructions for recording data on the Unknown and QA voucher spreadsheets. For example, a list of
growth-habit codes as well as floras and field guides are included for quick reference while other tabs
provide examples and specific instructions on how to fill out the various data fields of the Excel
spreadsheets for the QA voucher and Unknown specimen spreadsheets.
The identification spreadsheets were forwarded to the NWCA Data Management and Analysis Teams. The
Vegetation Analysis Team reviewed the identification spreadsheets submitted by the labs and
standardized nomenclature to the USDA-NRCS PLANTS database (USDA, NRCS 2019-2020). The validated
identifications of unknown taxa were integrated with the NWCA raw plant data tables, replacing the
pseudonyms recorded by the Field Crews for unknowns with their accepted scientific names (see Section
7.4.2).
7.4 Data Preparation - Parameter Names, Legal Values, and Data Validation
7.4.1 Description of Vegetation Field Data Tables
The data from the completed vegetation field forms were electronically submitted into several
predefined long format, raw data tables in the NWCA database. A separate table was created for each of
the three primary vegetation data forms:
• tbIPLANT - data from NWCA 2021 Vascular Species Presence and Cover (V-2)
• tbIVEGTYPE table - data from NWCA 2021 Veg Types/Ground Surface Attributes (V-3),.
• tbITREE table - data from NWCA 2021 Snags and Tree Counts (V-4)
Data from the NWCA 2021 Vascular Species Presence and Cover (V-2) form describe vascular plant
species identity, presence, cover, and height for each observed taxon and were collected in each 100-m2
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Veg Plot. Taxa typically represent species or lower level (e.g., subspecies, variety) classification, but
occasionally individual taxa were identified only to genus, family, or growth form. For convenience, in this
report, vascular plant taxa are generally referred to as species even though in some cases lower or higher
taxonomic levels are reflected. Form V-2 data used in candidate metric development for NWCA analyses
included taxon name, presence, and percent cover.
Other species level data were collected using Form V-2 but were reserved for further research and not
incorporated in the analysis of condition for NWCA. These other data included predominant height for
each species across each plot, and presence of individual species at different spatial scales (i.e., within 1-
m2 quadrats or 10-m2 quadrats) nested in the two corners of plot and within the overall 100-m2 plot. The
former can reflect vegetation structure and, when used with cover, volume by species or guild groups.
The latter address fine scale diversity patterns.
Data from the NWCA 2021 Veg Types/Ground Surface Attributes (V-3) form encompass descriptors of the
structure of vascular vegetation, non-vascular groups present, and ground surface attributes which are
each sampled in the five 100-m2 Veg Plots. All these data were used in developing candidate metrics.
Data from the NWCA 2021 Snags and Tree Counts (V-4) form include counts by diameter class of dead
trees/snags, as well as cover by height classes and by diameter classes for individual tree species in each
100-m2Veg Plot. Tree data were used in candidate metric development.
Parameter names and legal values or ranges for the field collected vegetation data are listed in Section
7.11, Appendix B. The quality of all the vegetation field data was carefully examined during data
validation.
7.4.2 Data Validation
Whenever large quantities of data are collected, it is not surprising for errors related to data or sample
collection, recording, sample analysis, or data entry to occasionally occur. Therefore, a series of quality
assurance (QA) review checks were conducted to identify and resolve any errors to ensure high quality
data. The NWCA established numerous cross-checks in the data collection and processing procedures,
within the protocols and field forms, to help limit potential errors during data collection. Verification and
update of the scanned vegetation data involved several QA steps conducted by members of the
Information Management Team and the Vegetation Analysis Team. Some checks required manual
evaluation of the paper forms or data scanned into the databases; others involved the use of specific R
Code written to identify records with specific kinds of potential errors. Tasks conducted by the
Information Management Team and the Vegetation Analysis Team are listed below.
Information Management Team:
• Verified that the data from the Vegetation Forms was submitted properly
• Where possible, verified spelling of plant species name with USDA PLANTS database
• Conducted quality assurance checks for valid ranges and legal values for all data
Vegetation Analysis Team:
• Updated the names for unknown taxa at each site based on plant specimen identification
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
• Conducted nomenclatural resolution, correcting species name spelling errors and resolving taxon
names that were recorded as synonyms to accepted names of the USDA PLANTS database (see
Section 7.5)
• Reviewed and resolved all instances of missing, out of range or non-legal values identified by the
IM Team:
o Review of the field forms often indicated a recording error that was readily resolved and
the data updated
o Where no resolution was apparent the data were flagged, and the error described
• Conducted logic checks and data type specific checks using R code to identify:
o Missing data (e.g., checking that if a certain type of data is present, another specific type
must also have a value)
o Recording errors (e.g., data recorded in a form workspace, rather than in the data field)
o Incongruities in values among related data
o Instances of individual plant species recorded multiple times at one site (i.e., multiple
data rows for the same species at one site which may have resulted when an unknown
was identified and was the same taxon as one already recorded)
• Determined the cause of each instance of a potential error revealed by logic checks
o Resolved these issues and provided a brief explanation of the issue and resolution in
tracking spreadsheets
• For all these data the relevant updates to the database were implemented using R-code, and a
brief explanation of the resolution was included with each of these records in the database
• For situations, where no resolution was apparent the data were flagged, and the errors described
The vast majority of concerns identified by these QA screenings were readily resolved allowing accurate
updates to the data. For the instances where specific issues could not be corrected, the data were flagged
with restrictions for use. Where corrections were needed, all original data values were retained as
inactive records in the NWCA database.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.5 Nomenclatural Standardization
Across the 2011, 2016 and 2021 field sampling
seasons, over 200 regional floras and field guides
were used by Field Crews for identification of
plants. Thus, a wide range of taxonomies were
applied to the occurrences of taxon-site pairs
observed across the United States. Consequently, a
critical step in data preparation was
standardization of plant nomenclature to ensure
that each taxonomic entity was called by the same
name throughout the NWCA study area. The
PLANTS nomenclatural database (USDA-NRCS
2020) was used as the national standard for
taxonomy for the NWCA.
In the NWCA 2021, plant species names originated
from raw data records collected using the Vascular
Species Presence and Cover (V -2) form, and from
lab identifications of unknown taxa that were
collected in the field. The process for reconciliation
of nomenclature outlined in Section 7.5.1 was used
for both data types. Section 7.5.2 provides a brief
description of procedures for taxonomic review
and documentation of name assignments that
were used for data from Form V 2. The
documentation process for the lab identifications
of unknown plants were similar but tailored to the
structure of these data.
Nomenclatural standardization was a complex
undertaking, and in this section, we provide an
overview of the process used for NWCA 2021.
7.5.1 Nomenclature Reconciliation Methods
We reconciled names for the NWCA 2021 observed plant taxa, at each location of their occurrence, to
the PLANTS nomenclatural database (hereafter, PLANTS) (USDA-NRCS 2020) using the methods (Figure
7-5) we developed for the 2011 and 2016 NWCA (USEPA 2023a). A series of automated filters based on
the components in Figure 7-5 were employed via R code, written using R software (R Core Team 2018-
2019), to link recorded names for NWCA observations to PLANTS accepted names and to identify names
and records 1) that matched accepted PLANTS names and 2) those that required further evaluation by a
botanist to resolve nomenclature.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Step 1: Identify NWCA name-location pairs directly matching PLANTS accepted names
A large proportion of the plant name-site pairs recorded in the NWCA could be directly matched to
PLANTS accepted names. These included records where:
1) The original NWCA name was the same as the accepted PLANTS name and there were no
synonyms for the name.
2) The original NWCA name pointed to one or more synonyms that all pointed to the same, single
accepted PLANTS name.
Step 2: identify NWCA name-location pairs needing botanical review to reconcile to PLANTS accepted
names
Even though most NWCA names could be directly matched to PLANTS nomenclature in Step 1, a large
number required botanical review to select the correct PLANTS accepted name. There were three
primary types of name issues which necessitated further botanical review:
1) Unmatched Names- no PLANTS accepted name or synonym matched a particular NWCA name-
site pair. Common reasons for unmatched names were misspelling or mis-scanning of the record
or use of an abbreviation or common name. Rarely, the taxon represented a name or taxon not
included in the PLANTS database.
2) Same Name with Different Authorities (shorthand terminology = Multiple Authorities) - refers to
an NWCA name which pointed to synonyms with exactly the same genus and species epithets,
but which had different botanical authorities for the name.
3) Species Concept Unclear - NWCA binomial name was contained in multiple potential PLANTS
accepted names or multiple synonym names that point to multiple possible PLANTS accepted
names.
Step 3: Review name-site pairs identified in Step 2 and determine correct name assignment
The set of names and records identified as requiring further evaluation were reviewed by the NWCA lead
botanist/ecologist, using a general stepwise procedure for nomenclatural determination:
1) Identify and correct spelling errors or abbreviated names.
2) Identify all synonyms and accepted PLANTS name(s) that could apply to each ambiguous taxon-
site pair name.
3) Compare geographic distribution of potential synonyms and accepted PLANTS names with
location of the observed NWCA taxon.
4) Review field records and notes from the NWCA Field Crew regarding the observed NWCA taxon.
5) Review the species concept for the taxon based on flora(s) used by field botanist, as well as other
pertinent taxonomic resources and floristic databases.
The procedures in Step 3 allowed determination of the PLANTS nomenclature accepted name for the
majority oftaxon-site pairs that needed botanical review. Fortaxa where the appropriate PLANTS
accepted name could not be definitively resolved using these procedures, a taxonomist at the PLANTS
database was consulted for assistance with final name determination. This consultation involved
discussions between the NWCA lead botanist/ecologist and the PLANTS taxonomist to review floras,
historical records, and floristic/taxonomic databases pertinent to each taxon-location pair considered. In
a few cases, the PLANTS taxonomist consulted with other botanists across the US with specific expertise
regarding a particular taxonomic group (e.g., species, genus, family) to resolve a naming issue.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Figure 7-5. Process for screening and reconciling names of plant taxa observed in the NWCA. Dark blue boxes = steps completed using R code, light blue boxes
= steps requiring botanical review, purple boxes = type of name resolution applied, and the dark blue central box = final name resolution.
84
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.5.2 Nomenclature Standardization Results and Documentation
For the 2011 NWCA, we developed and applied a standard approach for organizing, resolving, and
documenting the name reconciliations for plant name-site pairs needing review (USEPA 2016a). We used
the same general procedure to reconcile plant nomenclature to the PLANTS database for subsequent
NWCA surveys in 2016 and 2021. The NWCA 2011 plant data were also reviewed during the NWCA 2016
reconciliation process to identify and update any plant names that were no longer congruent with the
PLANTS database nomenclature(USDA-NRCS 2020).
Specific NWCA species records (including name, cover value, and other data), along with information
from the PLANTS database, were exported into an Excel Workbook for Nomenclature Resolution. This
workbook gathered key information in one location to facilitate review of the taxonomy and to highlight
when other information was needed. Important NWCA data elements included in the Excel Workbook
were NWCA SITEJD and UID, state, county, latitude and longitude, wetland type, a list of the floras used
by the Field Crew at a particular site, and a link to the scanned field form image. Access to the scanned
field form allowed easy viewing of any notes Field Crews may have made in relation to a particular
species, as well as a view of other taxa present at a site. Critical information from the PLANTS database
included synonyms and accepted names that could potentially correspond to the specific taxon-site pairs.
Various other location pertinent floristic resources and databases were also used when needed by the
NWCA botanist/ecologist in resolving name issues.
The Excel Nomenclature Workbook includes separate spreadsheet tabs for reviewing unresolved names
in three categories: Unmatched Names, Multiple Authorities, and Species Concept Issues (see Step 2 in
Section 7.5.1, for definitions). For each taxon-site pair to be evaluated (rows in spreadsheets) listed, the
associated columns (e.g., NWCA data, taxonomic and distributional information from the PLANTS
database, and other information) informed name resolution. An instruction page in the Workbook
described the associated data included in each of the spreadsheets and the ways this information could
aid in name determination. During nomenclatural review, the rationale for assignment of the correct
PLANTS accepted name to each name-site pair in the NWCA data was documented by specifying a reason
code and, where needed, providing narrative notes and citations of taxonomic sources.
Following taxonomic standardization, the master list of plants observed across all 3 survey years included:
5,947 taxa that occurred as 27,196 taxon-state pairs and 96,014 taxon-site pairs. The majority of taxa
observed in the NWCA were identified to the species, subspecies, or varietal level (n = 5,421, 28 of these
were hybrids). The remaining taxa in the list represented identifications made at higher taxonomic levels,
e.g., genus, family.
Once nomenclature for the NWCA name-site pairs was resolved, the appropriate accepted PLANTS name
was applied to each NWCA record. The original names recorded by the Field Crew or lab identifications
were retained as inactive data. Names (NWCA_NAME) and symbols (ACCEPTED_SYMBOL) for the 5,947
taxa are listed in the plant taxa file (nwca21_PlantTaxa-data.csv3). The NWCA_NAME typically reflects an
accepted scientific name from the PLANTS nomenclature and ACCEPTED_SYMBOL typically reflects the
PLANTS accepted symbol. In a few cases (19 taxa), where an appropriate taxonomic concept was not
available in PLANTS, we determined the name from other sources and assigned an ACCEPTED_SYMBOL
preceded by the number 1. For these 19 taxa, the complete name with authorities for the relevant taxon
can be found in the SCI_NAME_AUTH column. Taxa that were identifiable only to growth form were
3 .csv files referenced throughout the vegetation chapters are available to download from the USEPA NARS website
(https://www.epa.gov/national-aquatic-resource-surveys/data-national-aquatic-resource-surveys).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
assigned an NWCAJNAME representing one of 36 detailed standardized growth habit/category
designations, each of which were connoted by an ACCEPTED_SYMBOL preceded by the number "2".
7.6 Species Traits - Life History: Growth-habit, Duration, and Plant Category
Traits reflecting species life history based on growth-habit, duration, and plant category for all vascular
taxa observed in the NWCA were downloaded from the PLANTS database (USDA NRCS 2020). This trait
information was used directly or summarized to reduce the number of classes in each life history group.
Life history designations for each taxon observed in the 2011, 2016 and 2021 NWCAs are included in the
plant taxa file (inwca21_Plant.Taxa-data.csv). Life history information was used in combination with
presence, frequency, and cover data for individual species to develop candidate metrics to summarize the
distribution and importance of life history traits across each sampled site (see Section 8.8: Appendix E).
7.6.1 Growth-Habit
Primary growth-habit types for
the plant taxa observed were
based on growth-habit
designations in the PLANTS
database (USDA-NRCS 2019-
2020).
In the PLANTS database,
individual species were
frequently identified as spanning
multiple growth-habit types. This
results in numerous combined
growth- habit: categories, each
often representing few taxa. To
facilitate data analysis, we
merged some of multiple
growth-habit groups from the
PLANTS database into larger
categories for the NWCA data
analysis (Table 7-1).
7.6.2 Duration
Duration or longevity for plants is
described by annual, biennial, and perennial life cycles. Some individual species may exhibit different
durations depending on growing conditions. Consequently, in addition to the individual duration classes,
a variety of mixed duration categories occur in the PLANTS trait database. To facilitate data analysis, we
merged some multiple type groups from the PLANTS database into larger categories (Table 7-2).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 7-1. Growth-habit categories, for species observed in the 2011, 2016 and 2021 NWCAs and used in analysis,
with a crosswalk to PLANTS database growth-habit designations. Capitalized Growth-habit Category Names are
used in calculation of Growth-habit metrics (see Section 8.8: Appendix E).
NWCA Growth-habit
PLANTS Database Growth-habit 'Designations' for NWCA Observed Species included in
NWCA Growth-habit Category
GRAMINOID
'Graminoid'; 'Subshrub, shrub, graminoid'; "Graminoid, shrub, subshrub"; 'Graminoid,
shrub, vine; subshrub'; 'Graminoid, shrub'; "Subshrub, shrub, graminoid'
FORB
'Forb/herb'; 'Forb/herb, shrub'; 'Forb/herb, shrub, subshrub'; 'Forb/herb, subshrub';
'Forb/herb, subshrub, shrub'
SUBSHRUB-FORB
'Subshrub, forb/herb'; 'Subshrub, forb/herb, shrub'; 'Subshrub, shrub, forb/herb'
SUBSHRUB-SHRUB
'Subshrub, shrub'; 'Shrub, forb/herb, subshrub'; 'Shrub, subshrub'; 'Subshrub'
SHRUB
'Subshrub, forb/herb, shrub, tree'; 'Shrub, tree'; 'Shrub', 'Tree, subshrub, shrub'
TREE-SHRUB
'Tree, shrub'; 'Tree, shrub, subshrub'; 'Tree, shrub, vine'
TREE
'Tree'
VINE
'Vine'; 'Vine, forb/herb'; 'Subshrub, forb/herb, vine'; 'Forb/herb, vine'; 'Vine,
herbaceous'; 'Vine, forb/herb'; 'Vine, forb/herb, subshrub'
VINE-SHRUB
'Vine, shrub1; 'Vine, subshrub'; 'Subshrub, vine'; 'Shrub, vine'; 'Shrub, forb/herb,
subshrub, vine'; 'Shrub, subshrub, vine'
NWCA Growth-habit
NWCA Growth-habit Category Combinations
HERB
GRAMINOID + FORB
SHRUB-COMB
SUBSHRUB-SHRUB + SHRUB
TREE-COMB
TREE-SHRUB + TREE
VINE-ALL
VINE + VINE-SHRUB
Table 7-2. Duration categories used in the NWCA analyses and a crosswalk to PLANTS database duration
designations for NWCA observed species. Capitalized Duration Category Codes are used in calculation of Duration
Metrics (see Section 8.8: Appendix E).
NWCA Duration
Categories
PLANTS Database Duration 'Designations' for NWCA Observed Species
ANNUAL
'Annual'
ANN_BIEN
(Annual-Biennial)
'Annual, biennial'; 'Biennial'; 'Biennial, an'
ANN_PEREN
(Annual-Perennial)
'Annual, biennial, perennial'; 'Annual, perennial'; 'Annual, perennial biennial'; 'Biennial,
perennial'; 'Biennial, perennial, an';
PERENNIAL
'Perennial'; 'perennial, an'; 'Perennial, annual'; 'Perennial, annual, biennial'; 'Perennial,
biennial'; 'Perennial, biennial, an'; Perennial, biennial, annual'
7.6.3 Plant Categories
Several major plant categories were considered in summarizing raw data to develop guild-based
candidate metrics. The categories assigned for individual NWCA vascular taxa based on PLANTS
database categories were:
• Dicot
• Monocot
• Gymnosperm
• Fern
• Horsetail
• Lycopod
• Quillwort
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.7 Species I raits - Wetland Indicator Status
Hydrophytic status for plants observed in the NWCA surveys was defined
using Wetland Indicator Status (WIS) (Table 7-3), WIS values for
individual species vary across 7 Wetland Regions (Table 7-4). A WIS
category was assigned for each laxon wetland region pair observed (see
nwca21_PlantWIS-data.csv). Most of the NWCA WIS assignments
originated from the National Wetland Plant List (NWPL) (USAGE 2016-
2018). However, the NWPL lacked WIS values for some NWCA taxon-
wetland region pairs. NWCA evaluated this subset of taxon wetland
region pairs and assigned WIS values (Table 7-3) as appropriate: 1) UPL
to OBL, 2) NOL (not on the NWPL and too little information to assign UPL
to OBL, but often occurring in moist locations), or 3) UND (undetermined
due to limited information).
Table 7-3. Wetland Indicator Status (WIS) definitions. OBL, FACW, FAC, FACU and UPL defined by Lichvar 2016,
NOL and UND defined by NWCA. These seven WIS Categories are used in calculating Hydrophytic Status Metrics
(Section 6.8: Appendix E). The Numeric Ecological Value (ECOIND2) for each indicator status (UPL to OBL) is used in
calculating indices describing the hydrophytic status of the vegetation at each sampled site.
Wetland Indicator Status
Numeric Ecological
(WIS)
Qualitative Description
Value (ECOIND2)
OBL - Obligate
Almost always occur in wetland
5
FACW - Facultative Wetland
Usually occur in wetlands, but may occur in non-wetlands
4
FAC - Facultative
Occur in wetlands and non-wetlands
3
FACU - Facultative Upland
Usually occur in non-wetlands, but may occur in wetlands
2
UPL- Upland
Almost never occur in wetlands
1
NOL - Not on National
Not on NWPL, but observed in NWCA wetlands under wet
Wetland Plant List
or moist conditions
UND - Undetermined
Wetland status is undetermined
Table 7-4. Wetland regions within which wetland indicator status for individual plant species are defined, and a
crosswalk between USACE codes and NWCA codes for these regions is provided.
Wetland Regions Map [US Army Corps of Engineers,
National Wetland Plant List Map (USACE 2016-2018)]
Wetland Region
USACE/NWPL
Wetland
Region Code
NWCA
Wetland
Region Code
(C0E_REG_ID)
Atlantic and Gulf Coastal
Plain
AGCP
CSTL_ PLAIN
Arid West
AW
ARID W
NONE
MW
Eastern Mountains and
AW GP
Piedmont
EMP
E MTNS
EMP
Western Mountains,
AGCP
Valleys, and Coast
WMVC
W MTS
Great Plains
GP
GT PLAINS
Northcentral and
Northeast
NCNE
NE
Midwest
MW
MIDWEST
88
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.7.1 Wetland Indicator Status Assignment Process
All taxon-wetland region pairs observed (n = 11,827) in the NWCA surveys were assigned a wetland
indicator status (WIS) category (Table 7-3): OBL- obligate (n = 2,472), FACW - facultative wetland (n =
2,401), FAC - facultative (n =1,972 ), FACU - facultative upland (n = 2,188), UPL - upland (n = 1,877), NOL
- not on NWPL list but considered by NWCA to occur in wetlands some of the time (n = 235), or UND -
undetermined (n = 682). The process used for making these category assignments is outlined in Steps 1
through 3 below. Step 4 explains how the origin of the WIS value for each NWCA taxon-wetland region
pair was documented.
Step 1: WIS available directly from NWPL for 7,972 species-region pairs -Where available, the WIS
category from the National Wetland Plant List (NWPL) (USACE 2016-2018) was assigned to the
corresponding NWCA taxon-wetland region pair. Assignments were made based on nomenclatural and
wetland region matches between the NWPL and observed NWCA taxa. The NWPL provides taxon names
as binomials (genus and species) only. WIS values were assigned to all NWCA names that were binomials
and direct matches to the NWPL names. Some NWCA names represented lower taxonomic levels (e.g.,
subspecies or varieties). For NWCA names with subspecies or variety designations where the genus and
species name matched a binomial on the NWPL, the NWPL WIS category for that binomial was assigned
to the NWCA taxon.
Step 2: WIS assigned from multiple sources for2,543 species-region pairs- Each NWCA taxon-wetland
region pair representing a taxonomic level of species, subspecies, variety, or hybrid and not assigned a
WIS category in Stepl was evaluated to determine whether a WIS category could be assigned. This was a
two-step process:
• Step 2a - First each of these NWCA taxon-wetland region pairs was evaluated to see if it was a
synonym for a binomial included on the NWPL. If so, the taxon was assigned an NWPL WIS
category following the procedures in Step 1 (n =694).
• Step 2b - The taxon-wetland region pairs in the species and lower taxonomic level group that
were not synonyms for taxa on the NWPL list (n = 1,849), were reviewed using a variety of
sources of ecological information (e.g., primary floras, distributional databases, and expertise of
the NWCA vegetation analysis team) to determine if a WIS category might reasonably be
assigned. Based on this review, each of these taxon-wetland region pairs was assigned a WIS
category with the majority assigned to the UPL and NOL categories:
o OBL (n = 28)
o FACW (n = 15)
o FAC (n = 17)
o FACU (n = 14)
o UPL (n = 1,541)
o NOL (n = 234)
Step 3: WIS assigned for 1,312 higher level taxon-wetland region pairs- Finally 1,312 taxon-wetland
region pairs that were identified only to growth form, family, or genus, or that were nonvascular plants
were assigned a WIS category. The few nonvascular taxon-wetland region pairs (n = 16) included in the
NWCA taxa list were classified as UND. Most NWCA taxon-wetland region pairs identified only to family or
growth habit were assigned an undetermined (UND, n = 262) WIS. Aquatic growth form-region pairs were
assigned OBL (n = 9) status. Genus-level taxon-wetland region pairs (n = 1,033) were evaluated as to
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
whether species in those genera for a given wetland region tended to predominantly occur in wetlands or
uplands. Genera that had species that were predominantly a particular WIS category were assigned that
category; those for which species spanned a wide range of categories were assigned UND:
• OBL (n = 160)
• FACW (n = 186)
• FAC (n = 160)
• FACU (n = 92)
• UPL (n = 31)
• UND (n = 404)
Step 4: Documentation of WIS Value Origin for11,827observedNWCA taxon-wetlandregion pairs-1 n
addition to the WIS assignment for each of the 11,827 NWCA taxon-wetland region pairs, a source or
reason for each assignment was included in the WIS_SOURCE column of the wetland indicator status trait
table (nwca21_PlantWIS-data.csv) to provide documentation of its origin. WIS_SOURCE codes,
definitions, and included WIS categories are provided below:
• NWPL: WIS value directly from NWPL [OBL, FACW, FAC, FACU, UPL]
• NWPL-BINOM: WIS value from binomial on the NWPL [OBL, FACW, FAC, FACU, UPL],
• NWPL-NOMEN: WIS value from NWPL synonym of NWCA_NAME [OBL, FACW, FAC, FACU, UPL]
• NWPL-UPL: no WIS value listed on NWPL, but likely UPL based on habitat descriptions [UPL]
• NWPL-ADJREG: WIS value from NWPL for the same taxon from an adjacent wetland region [OBL,
FACW, FAC, FACU]
• NWCA-WIS: NWCA assigned WIS value based on other wetland indicator information or habitat
descriptions from floras pertinent to region [OBL, FACW, FAC, FACU]
• NWCA-EPIPAR: Taxon-wetland region pair is an epiphyte or parasite that occurs on wetland
species [NOL]
• NWCA-MOIST: Habitat descriptions indicate that taxon-wetland region pair often occurs under
moist to wet conditions [NOL]
• NWCA-GENUS: UPL- OBL assignment based on predominant wetland status for species in a genus
for a wetland region, or if too little information was available or a wide range of WIS values were
present in the genus it was assigned UND. [OBL, FACW, FAC, FACU, UPL, UND]
• NWCA-NI: Insufficient information for a WIS assignment for taxon-wetland region pair [UND]
• NONVASCULAR: 16 nonvascular taxa that were included non NWCA taxa list [UND]
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.8 Species I raits - Native Status
The number, proportion, or abundance
of native vs. nonnative flora at a given
location can help inform assessment of
ecological condition and stress (Magee
et al. 2019b). To calculate metrics
describing native and nonnative
components of the flora, it was first
necessary to determine the native status
of the vascular plant taxa observed in the
NVVCA (USEPA 2016a, Magee et al.
2019b). Here, the state-level native
status was determined for the
approximately 27,000 taxon-state pairs
observed in the 2011, 2016 or 2021
NWCA surveys across various states of
the conterminous US.
Assigning state-level native status for such a large number of taxon-state pairs across the scale of the
NWCA was a demanding task. First, there is currently no comprehensive national standard for native
status of plant species at the local or state level. Next, existing native status designations can be
ambiguous, and the understanding of indigenous species distributions is incomplete. In addition, defining
the concepts for native and nonnative is not always straightforward. Nonnative species may originate
from other countries or continents. Some species are native in one part of the United States, but
nonnative in another. Other taxa may both have alien and native components (e.g., genotypes,
subspecies, varieties, or hybrids).
Consequently, our first step in determining native status for the observed taxa-state pairs was to define
several concepts describing native status for the NWCA (Table 7-5).
Table 7-5. Definition of state-level native status designations for NWCA taxon-state pairs.
Native Status Designations
Native (NAT): Indigenous to specific states in the conterminous US
Alien fALIEN): Introduced + Adventive
Introduced (INTR): Indigenous outside of, and not native in, the conterminous US
Adventive (ADV): Native to some areas or states of the conterminous US, but introduced in the
location of occurrence
Cryptogenic (CRYP): Includes both Native and Alien genotypes, varieties, or subspecies
Undetermined (UND): Taxa identified at level of growth form, most families, or genera with both native
and alien species
Definitions from Magee et al. 2019b
Note: NWCA defines nonnative plants to include both alien and cryptogenic taxa (Magee et al. 2019b)
Cryptogenic species include taxa with both introduced (often aggressive) and native (generally less
prevalent) genotypes, varieties, or subspecies. Many cryptogenic species are invasive or act as ecosystem
engineers (Magee et al. 2019b), so we grouped them with alien species and considered them nonnative
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
for the purpose of indicating ecological stress. For example, see the Nonnative Plant Indicator (NNPI,
Chapter 10:) and metrics ending in AC (Section 8.8, Appendix E).
Using the definitions from Table 7-5 to determine state-level native status for each of the NWCA taxon-
state pairs, we reviewed existing native status designations for all NWCA taxon-state pairs from a variety
of taxonomic and ecological sources:
1) Floristic Databases (state and national levels)
2) State and Regional Floras and Checklists
3) PLANTS Database (USDA, NRCS 2020): Native status and species distribution (conterminous US)
4) Consultation with the PLANTS nomenclatural team
Items 1 through 3 above were used in the primary review of native status for the NWCA taxon-state pairs
and included numerous floristic sources (> 85). Final NWCA native status assignments for individual
taxon-state pairs were based on the body of evidence from relevant reviewed sources. The native status
review process was conducted by the NWCA Lead Ecologist/Botanist and another member of the
Vegetation Analysis Team with strong botanical expertise. One key element of the review was to search
native status designations based on the NWCA accepted name (see Section 7.5) and where needed, on
its synonyms. Many native status determinations were clear-cut, but others were more complex and
required extensive review of distributions and floristic sources. For taxa with particularly complex origins,
the nomenclature team at the PLANTS Database provided input based on their expertise and access to
resources describing species distributions and first collections to inform native status designations.
Native Status determinations for NWCA observed taxa were made for all species-state pairs, and
wherever possible for taxa identified only to genus-state pairs. Family- and growth form-state pairs were
designated 'Undetermined'. The native status designations are compiled in the native status trait table
(nwca21_PlantNative-data.csv). The taxon-state pairs were distributed as Native = 21,958, Introduced =
3,029, Adventive = 128, Cryptogenic = 333, and Undetermined = 1,782. The distribution of native status
among taxon-state pairs are presented as percentages in Figure 7-6.
PERECNT OF TAXON -STATE PAIRS
¦ ADV BCRYP ¦ INTR « NAT BUND
ADV CRYP
<1% 1%
«
Figure 7-6. Distribution of native status among taxon-state pairs presented as percentages.
Native status was used in conjunction with validated field collected vegetation data and with other
species trait information to calculate numerous candidate metrics (Section 8.8: Appendix E).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.9 Species I raits - Coefficients of Conservatism
Coefficients of Conservatism (C-values, also called
CCs) describe the "tendency of individual plant
species to occur in disturbed versus near pristine
conditions. C-values for individual species are state
or regionally specific and scaled from 0 to 10.
• A C-value of 1 indicates a widespread
generalist species that thrives under
disturbed conditions.
• A C-value of 10 indicates a species that
occurs in specific habitats that are
minimally disturbed (i.e., largely
unaltered).
• For the NWCA, nonnative taxa were
assigned a C-value of 0.
C-values are the primary building blocks of 1)
floristic quality indices and 2) metrics describing
vegetation sensitivity or tolerance to disturbance. Coefficients of Conservatism (C-values) for individual
plant taxa in particular locations reflect a taxon's response to anthropogenic disturbance and its habitat
specificity. C-values are applied to taxa by state, region, or habitat, so the C-value for a particular species
often varies by location. Typically, C-values are assigned by panels of expert botanists/ecologists and have
proven to be powerful tools in describing vegetation condition.
Floristic Quality (FQ) indices can be stand-alone indicators of condition or used as a component of a
VMMI (e.g., see Section 9.4). Floristic quality describes the complement of plant species occurring at a
site, and is based on summarization of species-specific, state or regional Coefficients of Conservatism that
rank the responsiveness of each species to disturbance (Swink and Wilhelm 1979, Wilhelm and Ladd
1988). FQ indices have proven utility as indicators of wetland condition in many regions of the US (e.g.,
Lopez and I ennessy 2002, Cohen et al. 2004, Bourdaghs et al. 2006, Miller and Wardrop 2006, Milburn et
al. 2007, Bried et al. 2013, Gara 2013, Bourdaghs 2014). Several kinds of FQ indices have been used to
describe wetland condition; the two most common are Mean Coefficient of Conservatism (Mean C) and
the Floristic Quality Assessment Index (FQAI). Both can be based on species presence only or weighted by
species abundance.
Sensitivity and tolerance to disturbance are often key attribute categories used in MMIs for other
biological assemblages and for some wetland VMMIs. For plants, sensitivity can be described based on
presence or abundance of taxa with high C-values, whereas tolerance may be based on presence or
abundance of taxa with low C-values.
See Appendix E Metrics (Section 8.8) for description of metrics based on C-values and for details of their
calculation. Several versions of Floristic Quality Assessment Index (FQAI) and of Mean Coefficient of
Conservatism (Mean C) were investigated in NWCA analyses as potential metrics for inclusion in one or
more Vegetation VMMIs. Metrics describing sensitivity and tolerance to disturbance were also screened
as of potential VMMI components.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
C-values for individual plant species were not universally available for all states or regions, nor for all taxa
observed across the 2011, 2016 and 2021 NWCAs. In addition, existing state or regional C-value lists were
not compiled together in a readily accessible format. Thus, to use C-values as a plant trait and calculate C-
value based metrics for the NWCA, it's necessary to obtain or develop state or regional C-values for the
plant taxa observed. A unique C-value was needed for each observed NWCA taxon-region pair
representing a specific plant taxon in either a specific state or a specific region within a state.
This required:
• Step 1 - Compiling and standardizing existing State and Regional C-Value Lists from across the
Conterminous US (Section 7.9.1)
• Step 2 - Assigning existing C-values, where available, to each taxon-region pair observed in the
2011, 2016 and 2021 NWCA surveys (Section 7.9.2)
• Step 3 - Developing C-values for each NWCA taxon-region pair observed for which there was no
existing C-value (Section 7.9.3)
• Step 4 - Finalizing NWCA C-value trait table (Section 7.9.4)
The final C-value assignments for the taxon-region pairs observed are located in the NWCA C-value Trait
Table (nwca21_PlantCval-data.csv) on the NWCA website.
7.9.1 Compilation of Existing State and Regional C-Value Lists from Across the
Conterminous US
An initial compilation of C-value lists (unpublished) was developed for the 2011 NWCA and is described in
the 2011 NWCA Technical Report (USEPA 2016a). C-value coverage for the western states was sparse;
consequently, USEPA convened an expert panel to assign C-values to NWCA taxon-state pairs observed in
the 2011 and 2016 NWCAs and occurring in Arizona, California, Idaho, New Mexico, Nevada, Oregon,
Texas, and Utah (Fennessy & Great Lakes Environmental Center Inc., 2019, unpublished). These two sets
of C-value lists served as the starting point for an updated, more extensive compilation of C-value lists.
For the 2016 NWCA analysis, the NWCA vegetation analysis team developed a standardized compilation
of C-value lists available at the end of 2019, and applicable to plant taxa occurring in specific individual
states or regions across the conterminous US. This unpublished compilation is hereon referred to as the
Compiled C-value Lists (unpublished draft) or the CCL. For the NWCA 2021 analysis, the vegetation
analysis team supplemented the CCL with recently published C-value lists. Citations for the individual C-
value lists included in the CCL and recently published lists are provided in Appendix D (Section 7.13).
The CCL ultimately contained C-values for over 124,000 taxon-region pairs, which were standardized for
potential use with observed NWCA taxon-region pairs. The 124,000+ taxon-region pairs in the CCL were
recorded under the parameter name C-VALUE_NWCA_USE and accompanying each of these taxon-region
pair C-values was the source abbreviation (see Appendix D, Section 7.13) for the specific C-Value List from
which it originated.
Because diverse approaches to list organization, data formats, and taxonomy were used across the
various C-value lists, it was necessary to standardize a variety of elements in the CCL. This standardization
was reflected in the C-values listed under C-VALUE_NWCA_USE so they could later be applied to
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
observed NWCA taxon-region pairs. The original C-value for each these records was also retained in the
CCL.
C-values in the C-VALUE_NWCA_USE field of the CCL were standardized as indicated in the bullet points
below:
• Standardization of nomenclature - The component C-value lists within the CCL used diverse
taxonomic nomenclatures, with scientific names for plant taxa derived from state or regional
floristic resources. To ensure that each taxonomic entity was referred to by the same name and
compatible with NWCA accepted names (see Section 7.5), taxonomy for the Compiled C-value
Lists was standardized, wherever possible, to the PLANTS database (USDA, NRCS 2019)
nomenclature. For each taxon-region pair, both the PLANTS name and the name or names from
the original C-value list were included in the CCL. When two or more synonyms for a single taxon-
region pair were subsumed under a single PLANTS name, a decision tree (see B in the text box in
Figure 7-7) was used to select among the C-values for the synonyms to apply for the NWCA
taxon-region pair based on the PLANTS name.
• Selecting C-value when multiple values were available for a taxon-region pair- In the CCL, there
were sometimes multiple C-values lists available for the same state or region. Consequently,
there could potentially be more than one C-value available for the same taxon in that
state/region. Where this occurred, a decision-tree (see A in the text box in Figure 7-7) was used
to choose the most update-to-date or most rigorous/comprehensive list source from which to
select the C-value for a specific taxon-region pair.
• Standardization of C-value formats -The methods and formats used for presentation of C-values
varied among the state and regional lists, with C-values sometimes expressed as whole numbers
ranging from 0 to 10 and sometimes as decimal numbers, e.g., 2.1, 6.7. Consequently, NWCA
standardized all C-values as whole numbers between 0 and 10. C-values originally expressed as
decimals were rounded to the nearest integer; for example, a C-value of 5.5 or higher was
rounded to 6.
• Standardization of C-value scoring for nonnative plant species - States and regional C-value lists
did not treat alien plant species uniformly. Some included nonnative species and others did not.
Among those that did, the methods used to assign C-values for alien species were not
standardized. For example, many states assigned a C-value of zero to all nonnative taxa, but
occasionally alien taxa were ranked on a gradient of invasiveness using a range of negative
integers for C-values to indicate increasing potential impact. To address this issue, NWCA
standardized C-values for taxon-region pairs indicated as nonnative species by the CCL to zero.
• Native taxon-region pairs listed in the CCL without C-values - were designated undetermined
('UND').
Finally, we note that the specific criteria for C-value assignment varies somewhat across different state or
regional lists, and this is likely to introduce some variability in C-values for taxon-region pairs listed in the
CCL that is not strictly related to taxon responses to disturbance or natural conditions. However, floristic
quality metrics calculated from C-values tend to be robust to many sources of noise.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Decision Tree for selecting C-value for potential use in the NWCA (C-VALUE_NWCA_USE)) when more than
one C-value existed for a taxon-region pair in the Compiled C-value Lists (unpublished)
Definitions:
• C-VALUE_NWCA_USE - C-value for a taxon-region pair that has been standardized for potential use with
observed NWCA taxon-region pairs
• 2011NWCA_CVALUE - interim C-values used in 2011 NWCA analysis for observed taxon-region pairs
(nwca2011_planttaxa_cc_natstat.csv on NWCA website)
• SOURCE 1 - for a state or region: 1) only one list exists, or 2) the oldest list if two lists exist
• SOURCE2 - for a state or region: Where two C-value lists exist for a state or region, the most recently
completed one.
A. Decision points for selection of taxon-region pair for which the accepted PLANTS name in the CCL relates
to only one taxon name from a C-value list(s) for a state or region
1. Are there multiple C-values for the taxa-location pair?
a. NO -> Use the available C-value
b. YES -> 2.
2. How many and which sources are there?
a. Only one source is available (2011NWCA_CVALUE1 or Source l2 or Source 23) -> Use the
available C-value
b. 2011NWCA_CValue and Source 1 -> 3.
c. Source 1 and Source 2-> 4.
d. 2011NWCA_CValue, Source 1 and Source 2 -> 5.
3. Are 2011NWCA_CValue and Source 1 equal?
a. NO -> Prioritize Source 1 where available. Use 2011NWCA_CValue when there is no Source 1
value, or Source 1 has no value for a specific taxon.
b. YES -> Use matching C-value
4. How do Source 1 and Source 2 compare?
a. Source 1 and Source 2 are equal -> Use matching -value
b. Source 1 and Source 2 differ by only one value -> Use Source 2
c. Source 1 and Source 2 differ by more than one value -> Review & decision by NWCA lead
botanist
d. Source 1 and Source 2 disagree on Nativity -> Review & decision by NWCA lead botanist
5. 2011NWCA_CValue, Source 1 and Source 2 are all available
a. Use Source 1 or Source2-> 4
B. Decision points where the accepted PLANTS name is applied to two to several names from an original C-
value list (synonyms of the PLANTS name)1
1. If C-values among synonyms for the accepted name differ by 2 values or less -> Choose the higher C-
value as the C-VALUE_NWCA_USE
2. If C-values among synonyms differ by more than 2 values -> Review and C-value decision by NWCA
lead botanist
3. If there is a difference in native status between listed synonym names -> Review and C-value decision
by NWCA lead botanist
1Note: The steps listed in B were completed for all taxon-region pairs observed in the 2011 and 2016 NWCAs
that occurred in the CCL, but due to time limitations may not have been completed for all records in the CCL
Figure 7-7. Text box outlining C-value selection decision tree when multiple C-values were available for one taxon-
region pair
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.9.2 Assigning Existing C-values to Taxon-Region Pairs Observed in the NWCA Surveys
Each available C-value list included in the CCL was assigned a geography (GEOG) which reflected the areas
to which that list was applicable, typically this was an individual state, or an EPA Level III Ecoregion
(USEPA 2013) falling within part of one or more adjacent states. To facilitate assigning C-values from the
Compiled C-value Lists (CCL) to observed NWCA taxa, each NWCA site sampled was assigned to an NWCA
C-value Regions (see NWCA_CREG16, in the NWCA site information files). Individual NWCA C-value
Regions were defined as one of the following: an entire individual state, the portion of a state described
by a specific Level III Ecoregion, or, in one case, the portions of a state falling on the east vs. west side of a
mountain divide. Thus, a given NWCA C-value Region could represent an entire state or a part of state.
Existing standardized C-values (see Section 7.9.1) were assigned to NWCA taxon-region pairs (where
region = NWCA C-value Region) with C-values for a particular region selected from one or more applicable
lists of the CCL. Often both a regional and a state C-value list were pertinent to a particular NWCA C-value
Region. In some instances, C-value coverage was incomplete for an NWCA C-value Region. When this was
the case, C-values were considered from nearby geographies, i.e., adjacent or nearby states with the
same or similar Level III Ecoregions. For each NWCA C-value Region, the NWCA lead botanist/ecologist
prioritized the applicable CCL lists in order of best geographic/ecoregional fit and availability of C-values.
R-code was developed to assign C-values from the CCL to taxa in each NWCA C-value Region based on the
prioritized order of the specific applicable C-value lists. Each NWCA C-value Region had two to four CCL
geographies, from which C-values could be drawn. The CCL geographies (and their accompanying C-value
lists) that were applicable to each NWCA C-value Region were given Priority 1, Priority 2, Priority 3, or
Priority 4 for order of use. C-values were assigned from the CCL to the NWCA taxon-region pairs in each
NWCA C-value region using the following order:
• C-values were assigned first from the Priority 1 List.
• If no C-value for a taxon-region pair was available in the Priority 1 List, then C-value was assigned
from the Priority 2 List.
• This process was continued sequentially through lists of subsequent priority levels until all
available C-values from the CCL for relevant NWCA taxon-region pairs were assigned.
Using this approach, existing C-values from the CCL were assigned to nearly 29,000 NWCA taxon-region
pairs. Of these taxon-region pairs, most were species or lower taxonomic-levels (i.e., subspecies, variety,
hybrid). However, the CCL also included C-values for some genera, and, where available, they were
applied to NWCA genus-region pairs.
7.9.3 Defining C-values for NWCA Taxon-Region Pairs Where None Were Available
After applying existing C-values from the CCL, some NWCA taxon-region pairs still lacked C-values. These
taxon-region pairs fell into three groups:
• Group 1 - identified to species or lower taxonomic levels (e.g., species, subspecies, variety, or
hybrid), hereafter species-region pairs
• Group 2 - identified to only to genus
• Group 3 - identified to only to high-level taxonomic categories (e.g., subfamily, family, growth
form, or a few nonvascular taxa)
Group 1 - C-value Assignment for Species-Region Pairs-The NWCA species-region pairs lacking C-values
were evaluated to determine whether a recently published C-value list, or an existing C-value from a
proximate geography that was not previously identified in the priority C-value lists for a particular NWCA
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
C-value region might be available for use. Using this approach, the following steps were used in
identifying C-values for these NWCA species-region pairs:
• If a C-value was available from a recently published list, it was applied to the species-region pair.
• If a C-value was not available from a new list, but a relevant C-value in an adjacent state and the
same Level III ecoregion was available, then it was applied to the NWCA species-region pair.
• If no such value was available, but a C-value from a nearby state and similar ecoregion was
available, that C-value was selected.
• If multiple C-values from nearby geographies might apply, these were reviewed by the NWCA
botanist/ecologist and the highest value (if there were 2 potential C-values) or the median value
(if there were 3 or more potential C-values) was selected.
• If the NWCA species-region pair was considered introduced or adventive by NWCA, a C-value of 0
was assigned.
• If no C-value could be assigned, the NWCA species-region pair was assigned UND.
In all cases, the list in the CCL, or recently published source, from which the C-value assigned to a NWCA
species-region pair originated was noted in the final trait table (see Appendix D for source list
abbreviations). Where the C-value was derived from the median of several C-value source lists or was
otherwise assigned by the NWCA botanist/ecologist, the C-value source was noted as NWCA16 or
NWCA21.
Group 2 - Genus-Region Pair Assignments-The NWCA genus-region pairs that were not initially assigned
C-values from the CCL were evaluated in a two-step process to see if C-values could be developed. First, a
tentative C-value was assigned based on the median C-value for species in the genus and occurring in the
NWCA C-Region and also appearing on the priority 1, or priority 1 and 2 C lists in the CCL. Note, some C-
lists include the flora of a state, but others include only a subset of the flora (e.g., sometimes only
wetland species). The NWCA botanist/ecologist then reviewed these tentative genus-region C-values and
accepted or rejected them using BPJ supported by information for the genus in the NWCA C-value region,
e.g.:
• The number of C-values and species in the genus that the median C-value represented.
• How well was the genus represented on the C-value lists applicable to the NWCA C-value region?
To address this question, PLANTS database maps or relevant floras were consulted to evaluate
distribution of taxa in genus in the NWCA C-value region.
• Were nonnative species included in the genus, and if so, how many, and are they typically
widespread invaders?
• Based on this review decide whether to accept median C-value or assign as undetermined.
Record notes on decisions.
Group 3- High-Levei Taxa or Growth Forms -The taxon-region pairs in this group were assigned
undetermined C-value (UND).
7.9.4 Final NWCA C-value Trait Table
The last step in finalizing the NWCA C-value Trait Table was ensuring that all taxa designated as
introduced or adventive by the NWCA (see Section 7.8) received a C-value of zero.
The final C-value assignments for the taxon-region pairs observed in the 2011, 2016 and 2021 NWCAs are
located in the NWCA C-value Trait Table (nwca21_PlantCval-data.csv) on the NWCA website. The source
from the CCL and recently published lists from which each NWCA taxon-region pair C-value originated is
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
also noted in this table and designated by abbreviations defined in Section 7.13, Appendix D. C-values or
UND status defined by NWCA are denoted by NWCA16 or NWCA21 as the C-value source.
The NWCA C-value Trait Table includes C-values specific to 31,641 NWCA taxon-region pairs:
• 27,868 species-region pairs with C-values ranging from 0 to 10 (here species includes: species,
subspecies, varieties, or hybrids)
• 2,394 genus-region pairs with C-values ranging from 0 to 10
• 1,379 taxon-region pairs where C-value remained undetermined (UND)
o 757 of these were family level or higher, taxa identified only to growth form, and a
handful of nonvascular taxa
o 284 of these were genus-region pairs
o 338 of these were species-region pairs
The NWCA C-values were used in calculation of floristic quality indices (e.g., variations of FQAI and Mean
C) and metrics describing sensitivity and tolerance to disturbance. See Section 8.8, Appendix E Metrics for
a list of specific metrics. The NWCA adopted the standard practice of excluding taxon-region pairs with C-
values = UND taxa from calculations of metrics of floristic quality and of disturbance sensitivity or
tolerance. The NWCA taxon-region pairs with C-values = UND represented a very small proportion of
NWCA taxa observed across all sites, and where these occurred, they typically had low abundance, so
their exclusion was expected to have little impact on metric values.
7.10 Literature Cited
Bourdaghs M, Johnston CA, Regal RR (2006) Properties and performance of the Floristic Quality Index in
Great Lakes Coastal Wetlands. Wetlands 26: 718-735
Bourdaghs M (2014) Rapid Floristic Quality Assessment Manual, wqbwm2-02b. Minnesota Pollution
Control Agency (MPCA), Saint Paul, Minnesota
Bried JT, Jog SK, Matthews JW (2013) Floristic quality assessment signals human disturbance over natural
variability in a wetland system. Ecological Indicators 34: 260-267
Cohen MJ, Carstenn S, Lane CR (2004) Floristic quality indicies for biotic assessment of depressional
marsh condition in Florida. Ecological Applications 14: 784-794
Euliss NH, Mushet DM (2011) A multi-year comparison of IPCI scores for Prairie Pothole Wetlands:
Implications of temporal and spatial variation. Wetlands 31: 713-723
DeKeyser ES, Kirby DR, Ell MJ (2003) An index of plant community integrity: Development of the
methodology for assessing prairie wetland plant communities. Ecological Indicators 3: 119-133
Deimeke E, Cohen MJ, Reiss KC (2013) Temporal stability of vegetation indicators of wetland condition.
Ecological Indicators 34: 69-75
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Galatowitsch S, Whited D, Tester J (1999) Development of community metrics to evaluate recovery of
Minnesota wetlands. Journal of Aquatic Ecosystem Stress and Recovery (Formerly Journal of Aquatic
Ecosystem Health) 6: 217-234
Genet J (2012) Status and Trends of Wetlands in Minnesota: Depressional Wetland Quality Baseline.
Minnesota Pollution Control Agency, Saint Paul, Minnesota
Gara B (2013) The Vegetation Index of Biotic Integrity "Floristic Quality" (VIBI-FQ). Ohio EPA Technical
Report WET/2013-2. Ohio Environmental Protection Agency, Wetland Ecology Group, Division of Surface
Water, Columbus, Ohio
Lichvar RW, Banks DL, Kirchner WN, and. Melvin NC (2016) The National Wetland Plant List: 2016 wetland
ratings. Phytoneuron 2016-30: 1-17. Published 28 April 2016. ISSN 2153 733X
http://www.phytoneuron.net/
Lopez RD, Fennessy MS (2002) Testing the floristic quality assessment index as an indicator of wetland
condition. Ecological Applications 12: 487-497
Mack JJ (2007) Integrated Wetland Assessment Program. Part 9: Field Manual for the Vegetation Index of
Biotic Integrity for Wetlands, v. 1.4. Ohio EPA Technical Report WET/2004-9. Ohio Environmental
Protection Agency, Wetland Ecology Group, Division of Surface Water, Columbus, Ohio
MackJJ, Kentula ME (2010) Metric Similarity in Vegetation-based Wetland Assessment Methods.
EPA/600/R-10/140. US Environmental Protection Agency, Office of Research and Development,
Washington, DC
Magee TK, Ringold PL, Bollman MA (2008) Alien species importance in native vegetation along wadeable
streams, John Day River basin, Oregon, USA. Plant Ecology 195: 287-307
Magee TK, Ringold PL, Bollman MA, Ernst TL (2010) Index of Alien Impact (IAI):A method for evaluating
alien plant species in native ecosystems. Environmental Management 45: 759-778
Magee TK, Blocksom KA, Fennessy MS (2019a) A national-scale vegetation multimetric index (VMMI) as
an indicator of wetland condition across the conterminous United States. Environmental Monitoring and
Assessment 191 (SI): 322, doi: 10.1007/sl0661-019-7324-4.
https://link.springer.com/article/10.1007/sl0661-019-7324-4
Magee TK, Blocksom KA, Herlihy AT, &. Nahlik AM (2019b) Characterizing nonnative plants in wetlands
across the conterminous United States. Environmental Monitoring and Assessment 191 (SI): 344, doi:
10.1007/s 10661-019-7317-3. https://link.springer.com/article/10.1007/sl0661-019-7317-3
Mclntyre S, Lavorel S (1994) Predicting richness of native rare, and exotic plants in response to habitat
and disturbance variables across variegated landscape. Conservation Biology 8(2): 521-531
Mclntyre S, Lavorel S, Landsberg J, Forbes TDA (1999) Disturbance response in vegetation -- towards a
global perspective on functional traits. Journal of Vegetation Science 10: 621-630
100
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Milburn SA, Bourdaghs M, Husveth JJ (2007) Floristic Quality Assessment for Minnesota Wetlands.
Minnesota Pollution Control Agency, St. Paul, Minnesota
Miller SJ, Wardrop DH (2006) Adapting the floristic quality assessment index to indicate anthropogenic
disturbance in central Pennsylvania wetlands. Ecological Indicators 6: 313-326
Miller SJ, Wardrop DH, Mahaney WM, Brooks RP (2006) A plant-based index of biological integrity (IBI) for
headwater wetlands in central Pennsylvania. Ecological Indicators 6: 290-312
Mitsch WJ, Gosselink JG (2007) Wetlands. John Wiley & Sons, Hoboken, NJ
Quetier F, Thebault A, Lavorel S (2007) Plant traits in a state and transition framework as markers of
ecosystem response to land-use change. Ecological Monographs 77: 33-52
R Core Team (2018-2019) R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. (http://www.R-project.orR/)
Reiss KC (2006) Florida Wetland Condition Index for depressional forested wetlands. Ecological Indicators
6: 337-352
Ringold PL, Magee TK, Peck DV (2008) Twelve invasive plant taxa in in US western riparian ecosystems.
Journal of North American Benthological Society 27: 949-966
Rocchio J (2007) Assessing Ecological Condition of Headwater Wetlands in the Southern Rocky Mountains
Using a Vegetation Index of Biotic Integrity (Version 1.0). Colorado State University, Colorado Natural
Heritage Program, Fort Collins, Colorado
Rooney R, Bayley S, Rooney RC, Bayley SE (2012) Development and testing of an index of biotic integrity
based on submersed and floating vegetation and its application to assess reclamation wetlands in
Alberta's oil sands area, Canada. Environmental Monitoring and Assessment 184: 749- 761
Swink F, Wilhelm G ( 1979) Plants of the Chicago region: A Checklist of the Vascular Flora of the Chicago
Region,with Keys, Notes on Local Distribution, Ecology, and Taxonomy, and a System for Evaluation of
Plant Communities. Morton Arboretum, Lisle, Illinois
Tiner RW (1999) Wetland Indicators: A Guide to Wetland Identification, Delineation, Classification, and
Mapping. Lewis Publishers, Boca Raton, FL, USA
USACE (2016-2018) National Wetland Plant List, version 3.3. and 2018 NWPL Speice Update. US Army
Corps of Engineers Engineer Research and Development Center, Cold Regions Research and Engineering
Laboratory, Hanover, NH
http://wetland-plants.usace.army.mil/ [Accessed February 2020, downloaded: National_2016v3.xlsx and
2018_NWPL_Update_Species.xlsx]
USDA, NRCS. (2020). The PLANTS Database (http://plants.usda.gov). National Plant Data Team,
Greensboro, NC USA)
101
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
USEPA (2002) Methods for Evaluating Wetland Condition: #10 Using Vegetation to Assess Environmental
Conditions in Wetlands. Office of Water, US Environmental Protection Agency, Washington, DC
USEPA (2006) Wadeable Streams Assessment: A Collaborative Survey of the Nation's Streams. EPA 841-B-
06-002. US Environmental Protection Agency, Washington, DC
USEPA (2011a) National Wetland Condition Assessment: Field Operations Manual. EPA-843-R10-001. US
Environmental Protection Agency, Washington, DC
USEPA (2013) Level III ecoregions of the continental United States: Corvallis, Oregon, US Environmental
Protection Agency - National Health and Environmental Effects Research Laboratory, map scale
1:7,500,000, https://www.epa.gov/eco-research/level-iii-and-iv-ecoregions-continental-united-states.
USEPA (2016a) National Wetland Condition Assessment: 2011 Technical Report. EPA-843-R-15-006. .US
Environmental Protection Agency, Washington, DC. https://www.epa.gov/national-aquatic-resource-
surveys/national-wetland-condition-assessment-2011-results
USEPA (2016b) National Wetland Condition Assessment 2011: A Collaborative Survey of the Nation's
Wetlands. EPA-843-R-15-005. US Environmental Protection Agency, Washington, DC.
https://www.epa.gov/national-aquatic-resource-surveys/national-wetland-condition-assessment-2011-
results.
USEPA (2016c) National Wetland Condition Assessment 2016: Field Operations Manual. EPA-843-R-15-
007. US Environmental Protection Agency, Washington D.C.
USEPA (2016d). National Wetland Condition Assessment 2016: Laboratory Operations Manual. EPA-843-
R-15-009. US Environmental Protection Agency, Office of Water, Washington, DC.
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual (EPA-843-B-21-
002. US Environmental Protection Agency, Office of Water, Washington, DC.
USEPA (2021b) National Wetland Condition Assessment 2021: Laboratory Operations Manual (EPA-843-B-
21-003. US Environmental Protection Agency, Office of Water, Washington, DC.
USEPA (2023a) National Wetland Condition Assessment 2021: Technical Support Document. EPA-841-B-
23-001. US Environmental Protection Agency, Washington, DC.
Veselka W, Rentch JS, Grafton WN, Kordek WS, Anderson JT (2010) Using Two Classification Schemes to
Develop Vegetation Indices of Biological Integrity for Wetlands in West Virginia, USA. Environmental
Monitoring and Assessment 170: 555-569
Wilhelm G, Ladd D (1988) Natural Area Assessment inthe Chicago region. In: Transactions 53rd North
American Wildlife and Natural Resources Conference, Louisville, Kentucky. Wildlife Management
Institute,Washington, DC, pp 361-375
Wilson M, Bayley S, Rooney R (2013) A plant-based index of biological integrity in permanent marsh
wetlands yields consistent scores in dry and wet years Aquatic Conservation-Marine and Freshwater
Ecosystems 23: 698-709
102
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
7.11 Appendix B: Parameter Names for Field Collected Vegetation Data
PARAMETER
NAME
DESCRIPTION
RESULT
VALID RANGE/
LEGAL VALUES
Form V-l: NWCA 2016 Vegetation Plot Establishment
Plot Predominant Wetland Type Data: Observations from each of five 100-m2 (10x10m) Veg Plots
WETLAND TYPE
NWCA Target Wetland Type
dominating Veg Plot
One Category: EH - Estuarine Intertidal
Emergent, EW - Estuarine
Shrub/Forested, PRL-EM -Palustrine,
Lacustrine, or Riverine Emergent, PRL-
SS - Palustrine, Lacustrine, or Riverine
Scrub/Shrub, PRL-FO - Palustrine,
Lacustrine, or Riverine Forested, PRL-
UBAB- Palustrine, Lacustrine, or
Riverine Unconsolidated Bottom, PRL-f
- Palustrine, Lacustrine, or Riverine
previously farmed (not currently
actively farmed)
EH, EW,
PRLEM,
PRLSS,
PRLFO,
PRLUBAB, or
PRLF
Form V-2a and V-2b: NWCA Vascular Species Presence and Cover
Plant Species Data: Cover, presence, and height data for each vascular plant species observed in each of five 100-
m2 (10x10m) Veg Plots. Presence of each species in four component nested quadrats for each Veg Plot.
SPECIES
Scientific Name for each
species (taxon) encountered in
the Veg Plot.
Typically, the binomial genus and
species name. In some cases: lower
taxonomic levels (e.g., subspecies,
varieties) or higher taxonomic levels
(e.g., genus, family, growth form) or
pseudonyms for unknowns
Taxon name
SW
For each species present, the
smallest scale at which it is first
observed: l-m2or 10-m2
quadrat in SW corner or in
larger 100-m2 Veg Plot
One of: S = 1-m2 quadrat, M = 10-m2
quadrat, or W = the whole 100-m2
Veg Plot
S, M, orW
NE
For each species present, the
smallest scale at which it is first
observed: l-m2 or 10-m2
quadrat in NE corner or in
larger 100-m2 Veg Plot
One of: S = 1-m2 quadrat, M = 10-m2
quadrat, or W = the whole 100-m2
Veg Plot
S, M, orW
HEIGHT
Predominant height class for
each species present across a
Veg Plot
One Height Class: 1 = < 0.5m, 2 = >
0.5m-2m, 3 = > 2-5m, 4 = > 5-15m, 5
= > 15-30m, 6 = > 30m, or E = Liana,
vine, or epiphyte species
1, 2,3, 4, 5, 6, or
E
COVER
Percent cover of each species
across a Veg Plot
Cover value for each individual
species present is estimated as a
direct percentage of the spatial area
of the plot overlain by that species
and can range from 0 to 100%.
0-100%
Form V-3: NWCA Vegetation Types (Front) and Ground Surface Attributes (Back)
% Cover Vascular Vegetation Strata
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
PARAMETER
NAME
DESCRIPTION
RESULT
VALID RANGE/
LEGAL VALUES
SUBMERGED_AQ
% Cover Submerged Aquatic
Vegetation
0-100 % Cover
0-100%
FLOATING_AQ
% Cover Floating Aquatic
Vegetation
0-100 % Cover
0-100%
LIANAS
% Cover Lianas, vines, and vascular
epiphytes
0-100 % Cover
0-100%
Cover for other vascular vegetation in height classes indicated below:
VTALL_VEG
% Cover Vegetation > 30m tall
0-100 % Cover
0-100%
TALL_VEG
% Cover Vegetation > 15m to 30m
tall
0-100 % Cover
0-100%
HMED_VEG
% Cover Vegetation > 5m to 15m
tall
0-100 % Cover
0-100%
MED_VEG
% Cover Vegetation >2m to 5 tall
0-100 % Cover
0-100%
SMALL_VEG
% Cover Vegetation 0.5 to 2m tall
0-100 % Cover
0-100%
VSMALL_VEG
% Cover Vegetation < 0.5m tall
0-100 % Cover
0-100%
% Cover and Cateaorical Data for Non-Vascular Taxa
BRYOPHYTES
% Cover of Bryophytes growing on
ground surfaces, logs, rocks, etc.
0-100 % Cover
0-100%
PEAT_MOSS
Bryophytes dominated by
Sphagnum or other peat forming
moss
Y (yes) if present
Y
LICHENS
% Cover of Lichens growing on
ground surfaces, logs, rocks, etc.
0-100 % Cover
0-100%
ARBOREAL ABUN
Abundance of Arboreal Bryophytes
Categorical classes: ABUNDANT,
ABUNDANT,
DANCE
and Lichens
COMMON, SPARSE, NONE
COMMON,
SPARSE, or
NONE
ALGAE
% Cover of filamentous or mat
forming algae
0-100 % Cover
0-100%
MACROALGAE
% Cover of macroalgae (freshwater
species/seaweeds)
0-100 % Cover
0-100%
Water Cover and Depth
TOTAL_WATER
Total percent cover of water
across Veg Plot area
% Cover
0-100%
PREDOMINANT_D
Predominant water depth
depth in cm
Investigate if
EPTH
>200 cm
TIME
Time water depth measurements
were made
time on 24 hour clock
500 to 2100
(investigate if
outside this
range)
Bare around and Litter
Cover of bare ground = a + b + c < 100%
EXPOSED_SOIL
a) Cover exposed soil/sediment
% Cover
< 100%
EXPOSED_GRAVEL
b) Cover exposed gravel/cobble
(~2mm to 25cm)
% Cover
< 100%
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
PARAMETER
NAME
DESCRIPTION
RESULT
VALID RANGE/
LEGAL VALUES
EXPOSED_ROCK
c) Cover exposed rock (>25cm)
% Cover
< 100%
Vegetation Litter
TOTAL_UTTER
Total cover of vegetation litter
% Cover
< 100%
PREDOMINA NT_UTT
ER
Predominant litter type
G=Graminoid (e.g., grasses, sedges,
rushes), F=Forb, R=Fern,
C=Coniferous Tree/shrub,
D=Deciduous Tree/shrub,
E=Broadleaf Evergreen Tree/shrub
CONIFEROUS,
DECIDUOUS,
GRAMINOID,
FORB, FERN,
BROADLEAF
DEPTH_SW
Litter depth (cm) in center of 1-
m2 quadrat at SW corner of Veg
Plot
depth in cm
Investigate if
>100 cm
DEPTH_NE
Litter depth (cm) in center of 1-
m2 quadrat at NE corner of Veg
Plot
depth in cm
Investigate if
>100 cm
Cover of Downed Dead Woody Material (angle of incline < 45°)
WD_FINE
Cover of fine woody debris
(<5cm diameter)
% Cover
0-100%
WD_COARSE
Cover of coarse woody debris
(> 5cm diameter)
% Cover
0-100%
Form V-4a and V-4b: NWCA Snag and Tree Counts and Tree Cover
Standina Dead trees/snaas (<5cm DBH)
STANDING
Estimate of small standing
trees/snags on plot
Abundance Class: None (0), Few (1-
10), Common (11-20), Many (>20)
NONE, FEW,
COMMON,
MANY
Standina Dead Tree/Snaa Counts bv DBH Class
XXTHIN_SNAG
Dead trees/snags 5 to 10 cm
DBH (diameter breast height)
Counts
Investigate if >
200
XTHIN_SNAG
Counts of dead trees/snags 11
to 25cm DBH
Counts
Investigate if >
200
THIN_SNAG
Counts of dead trees/snags 26
to 50cm DBH
Counts
Investigate if >
100
JR_SNAG
Counts of dead trees/snags 51
to 75cm DBH
Counts
Investigate if >
20
THICK_SNAG
Counts of dead trees/snags 76
to 100cm DBH
Counts
Investigate if >
20
XTHICK_SNAG
Counts of dead trees/snags 101
to 200 cm DBH
Counts
Investigate if >
20
Tree Data
Tree Species Name
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
105
-------
PARAMETER
NAME
DESCRIPTION
RESULT
VALID RANGE/
LEGAL VALUES
TREE_SPECIES
Scientific Name for each tree
species (taxon) encountered in
the Veg Plot.
Typically, the binomial genus and
species name. In some cases: lower
taxonomic levels (e.g., subspecies,
varieties) or higher taxonomic levels
(e.g., genus, family, growth form) or
pseudonyms for unknowns
Taxon Name
Tree Species Cover by Height Class
VSMALL_TREE
For each tree species, cover of
trees < 0.5m tall
0-100 % Cover
0-100%
SMALL_TREE
For each tree species, cover of
trees 0.5m to 2m tall
0-100 % Cover
0-100%
LMED_TREE
For each tree species, cover of
trees > 2 to 5m tall
0-100 % Cover
0-100%
HMED_TREE
For each tree species, cover of
trees > 5m to 15m tall
0-100 % Cover
0-100%
TALL_TREE
For each tree species, cover of
trees > 15m to 30m tall
0-100 % Cover
0-100%
VTALL_TREE
For each tree species, cover of
trees > 30m tall
0-100 % Cover
0-100%
Tree Species Counts by DBH Class
XXTHIN_TREE
For each tree species, counts of
trees 5 to 10 cm DBH (diameter
breast height)
Counts
Investigate if >
200
XTHIN_TREE
For each tree species, counts of
trees 11 to 25cm DBH
Counts
Investigate if >
100
THIN_TREE
For each tree species, counts of
trees 26 to 50cm DBH
Counts
Investigate if >
50
JR_TREE
For each tree species, counts of
trees 51 to 75cm DBH
Counts
Investigate if >
20
THICK_TREE
For each tree species, counts of
trees 76 to 100cm DBH
Counts
Investigate if >
10
XTHICK_TREE
For each tree species, counts of
trees 101 to 200 cm DBH
Counts
Investigate if >
5
XXTHICK_TREE
For each tree species, counts of
trees > 200 cm DBH
Counts
Investigate if >
5
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
106
-------
7.12 Appendix D: Existing Coefficient of Conservatism Lists included in the
Compiled C-value Lists {unpublished draft) assembled by NWCA
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
All 48
Individual
Conterminous
United States
NWCA11
USEPA (2016) US Environmental Protection Agency. National Aquatic
Resource Surveys. National Wetland Condition Assessment 2011 (NWCA
2011 Plant CC and Native Status Values - Data (CSV) and NWCA 2011 Plant
CC and Native Status Values - Metadata (TXT)). [Includes (for observed plant
species) state-level trait information for: C-Values, Native Status
Designations, and Disturbance Sensitivity Categories], Available from USEPA
website: https://www.epa.gov/national-aquatic-resource-surveys/data-
national-aquatic-resource-surveys.
AZ
EPA19_AZ
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019, unpub.).
Project to assign C-values to western states for use in the USEPA National
Wetland Condition Assessment (NWCA): C-values for taxon-state pairs
observed in AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA
Surveys. Funded by USEPA Office of Wetlands, Oceans, and Watersheds to
the Great Lakes Environmental Center, Traverse City, Ml. EP-C-16-
008: Task Order #08. Unpublished Report and Excel File
(NWCA_C_Values Western States_ll-6-2018_Draft) Submitted to USEPA.
AZ
Armstrongl9
Armstrong, J, W. Hodgson, S. Jones, K. Barron, D. Setaro and A. B. Raschke
(2021). Floristic Quality Assessment: Development and Application in
Maricopa County, Arizona. Maricopa County Parks and Recreation and
Desert Botanic Garden.
CA
EPA19_CA
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019). Project to
assign C-values to western states for use in the USEPA National Wetland
Condition Assessment (NWCA): C-values for taxon-state pairs observed in
AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA Surveys.
Funded by USEPA Office of Wetlands, Oceans, and Watersheds to the Great
Lakes Environmental Center, Traverse City, Ml. EP-C-16-008: Task Order #08.
Unpublished Report and Excel File (NWCA_C_Values Western States_ll-6-
2018_Draft) Submitted to USEPA.
CO
ROCC_07
Rocchio, J. (2007). Floristic quality assessment indices for Colorado plant
communities. Fort Collins, Colorado: Colorado Natural Heritage Program,
Colorado State University.
CO
CNHP20
Smith, P., G. Doyle, and J. Lemly. 2020. Revision of Colorado's Floristic
Quality Assessment Indices. Colorado Natural Heritage Program, Colorado
State University, Fort Collins, Colorado.
CT
NEIW13_CT
New England Interstate Water Pollution Control Commission (NEIWPCC).
(2013) Northeast Regional Floristic Quality Assessment. Current URL (27
August 2019): https://neiwpcc.org/our-programs/wetlands-aquatic-
species/nebawwg/nqa/. Individual State CoC lists: Connecticut.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
DE
MCAV12
McAvoy, W.A. (2011) The Flora of Delaware Online Database. Delaware
Division of Fish and Wildlife, Natural Heritage and Endangered Species
Program, Smyrna, Delaware, http://www.wra.udel.edu/de-flora. Current
URL (28 August 2019): http://www.wrc.udel.edu/de-flora/
FL (Source 1)
LANE03
Lane, C.R., Brown, M., Murray-Hudson, M. and Vivas, M. B. (2003) The
Wetland Condition Index (WCI): Biological indicators for Isolated
Depressional Herbaceous Wetlands in Florida, Report Submitted to the
Florida Department of Environmental Protection under Contract #WM-683
FL (Source 2)
REIS05
A) Reiss, K.C.& Brown, M.T. (2005a) The Florida Wetland Condition Index
(FWCI): Developing Biological Indicators for Isolated Depressional Forested
Wetlands. B) Reiss, K.C.& Brown, M.T. (2005) Pilot Study - The Florida
Wetland Condition Index (FWCI): Preliminary Development of Biological
Indicators for Forested Strand and Floodplain Wetlands. Report Submitted to
the Florida Department of Environmental Protection Under Contract #WM-
683
FL_South
MORT09
Mortellaro, S., Barry, M., Gann, G., Zahina, J., Channon, S., Hilsenbeck, C.,
Scofield, D., Wilder, G., & Wilhelm, G. (2009). Coefficients of Conservatism
Values and the Floristic Quality Index for the Vascular Plants of South
Florida. Southeastern Naturalist, ll(mo3), 1-62, 62.
GA
ZOML13
Zomlefer, W. B., Chafing. L.G., Carter, J.R. and Giannasi, D.E. (2013)
Coefficient of Conservatism Rankings for the Flora of Georgia: Wetland
Indicator Species. Southeastern Naturalist 12:790-808.
IA
DROBOl
Drobney, P.D., Wilhelm, G.S., Horton, D., Leoschke, M., Lewis, D., Pearson, J.,
Roosa, D., and Smith, D. (2001) Floristic quality assessment for the state of
Iowa. Unpublished report.
ID
EPA19JD
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019). Project to
assign C-values to western states for use in the USEPA National Wetland
Condition Assessment (NWCA): C-values for taxon-state pairs observed in
AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA Surveys.
Funded by USEPA Office of Wetlands, Oceans, and Watersheds to the Great
Lakes Environmental Center, Traverse City, Ml. EP-C-16-008: Task Order #08.
Unpublished Report and Excel File (NWCA_C_Values Western States_ll-6-
2018_Draft) Submitted to USEPA.
IL
TAFT03
Taft, J.B., Wilhelm, G.S., & Masters, L.A. (2003) Floristic quality assessment
for vegetation in Illinois a method for assessing vegetation integrity. Illinois
Native Plant Society
IL
USACE17
Herman, B., Sliwinski, R. and S. Whitaker (2017). Chicago Region FQA
(Floristic Quality Assessment) Calculator. U.S. Army Corps of Engineers,
Chicago, IL.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
108
-------
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
IN
ROTH19
Rothrock, P.E. (2019) The Floristic quality assessment of Indiana concepts,
use and development of coefficients of conservatism. Final Report for ARN
A305-4-53 EPA Wetland Program Development Report Grant CD975586-01
IN
ROTH04
Rothrock, P.E. (2004) The Floristic quality assessment of Indiana concepts,
use and development of coefficients of conservatism. Final Report for ARN
A305-4-53 EPA Wetland Program Development Report Grant CD975586-01.
KS
FREE12
Freeman, C. C. (2012) Coefficients of Conservatism for Kansas Vascular
Plants (2012) and Selected Life History Attributes. Kansas Biological Survey,
University of Kansas, http://ksnhi.ku.edu/media/ksnhi/public-data-
resources/Coefficients%20of%20Conservatism%20for%20Kansas%20Vascula
r%20Plants%20%282012%29.pdf
KY
WHIT97
White, D., Shea, M., Ladd, D. and Evans, M. (1997) Floristic quality
assessment for Kentucky. The Kentucky Chapter of The Nature Conservancy,
Kentucky State Nature Preserves Commission, The Missouri Chapter of The
Nature Conservancy
LA
CRET12
Cretini, K.F., Visser, J.M., Krauss, K.W., & Steyer, G.D. (2012) Development
and use of a floristic quality index for coastal Louisiana marshes.
Environmental Monitoring and Assessment. 184:2389-2403. List included as
supplement to paper. An updated list with a few added species was provided
to Nicole Kirchner in July 2012.
LA
Allain06
Allain, L., L. Smith, C. Allen, M.F. Vidrine, and J.B. Grace (2006). A Floristic
Quality Assessment System for the Coastal Prairie of Louisiana. Proceedings
of the 19th North American Prairie Conference.
MA
NEIW13_MA
New England Interstate Water Pollution Control Commission (NEIWPCC).
(2013). Northeast Regional Floristic Quality Assessment. Current URL (27
August 2019): https://neiwpcc.org/our-programs/wetlands-aquatic-
species/nebawwg/nqa/. Individual State CoC lists: Massachusetts.
ME
NEIW13_ME
New England Interstate Water Pollution Control Commission (NEIWPCC).
(2013). Northeast Regional Floristic Quality Assessment. Current URL (27
August 2019): https://neiwpcc.org/our-programs/wetlands-aquatic-
species/nebawwg/nqa/. Individual State CoC lists: Maine.
ME
MENA14
Maine Natural Areas Program. (2014) Coefficient of Conservatism Scores for
Maine. Maine Natural Areas Program, Augusta, Maine, USA. Current URL (8-
28-2019: https://www.maine.gov/dacf/mnap/features/coc.htm
Ml
REZN14
Reznicek, A.A., Penskar, M.R., Walters, B.S. and Slaughter, B.S. (2014)
Michigan Floristic Quality Assessment Database. Herbarium, University of
Michigan, Ann Arbor, Ml and Michigan Natural Features Inventory, Michigan
State University, Lansing, Ml. (http://michiganflora.net/home.aspx)
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
109
-------
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
Ml
HERM01
Herman, K.D., Masters, L.A., Penskar, M.R., Reznicek, A.A., Wilhelm, G.S.,
Brodovich, W.W. and Gardiner, K.P. (2001) Floristic quality assessment with
wetland categories and examples of computer applications for the state of
Michigan. 2nd Edition. Michigan Dept. of Natural Resources, Lansing, Ml. 19
pp. + appendices.
http://www.michigandnr.com/publications/pdfs/HuntingWildlifeHabitat/FQ
A_text.pdf
MN
MILB07
Milburn, S. A., Bourdaghs, M. and Husveth (2007) Floristic Quality
Assessment for Minnesota Wetlands. Minnesota Pollution Control Agency,
St. Paul, Minn. Accessed at www.pea.state.mn.us/water/biomonitoring/bio-
wetlands.html
MO
LADD93
Ladd, D. (1993) Coefficients of conservatism for the Missouri vascular flora: a
database of the flora of missouri with species conservatism coefficients,
wetness ratings, physiognomy, standardized acronyms, and common names.
Missouri chapter of the Nature conservancy.
MO
LADD15
Ladd, D. and Thomas, J.R. (2015) Ecological checklist of the Missouri flora for
Floristic Quality Assessment. Phytoneuron. 2015-12: 1-274. Published 12
February 2015. ISSN 2153 733X
MS
HERM06
Herman, B.D., Madsen, J. D. and Ervin, G.D. (2006) Development of
coefficients of conservatism for wetland vascular flora of north and central
Mississippi. Mississippi State University, GeoResources Institute Report 4001
(Water Resources)
MT
PIPP15_16
Pipp, A. (2015) Coefficient of Conservatism Rankings for the Flora of
Montana: Part 1.
Report to the Montana Department of Environmental Quality, Helena,
Montana. Prepared by
the Montana Natural Heritage Program, Helena, Montana. 73 pp
MT
PIPP15_16
Pipp, A. (2016) Coefficient of Conservatism Rankings for the Flora of
Montana: Part II.
Report to the Montana Department of Environmental Quality, Helena,
Montana. Prepared by
the Montana Natural Heritage Program, Helena, Montana. 75 pp.
MT
PIPP17
Pipp, A. (2017) Coefficient of Conservatism Rankings for the Flora of
Montana: Part III. Report to the Montana Department of Environmental
Quality, Helena, Montana. Prepared by the Montana Natural Heritage
Program, Helena, Montana. 107 pp
MT
JONE05
Jones, W.M. (2005) A vegetation index of biotic integrity for small order
streams in southwestern Montana and floristic quality assessment for
western Montana wetlands. Report to the Montana Department of
Environmental Quality and US Environmental Protection Agency, Montana
Natural Heritage program, Helena Montana. 29 pp. plus appendices.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
ND
TNGP01
The Northern Great Plains Floristic Quality Assessment Panel. (2001)
Coefficients of conservatism for the vascular flora of the Dakotas and
adjacent grasslands: US Geological Survey, Biological Resources Division,
Information and Technology Report USGS/ BRD/ITR—2001-0001, 32 p.
NE
ROLF11
Rolfsmeier, S. & Steinauer, G. (2003, 2011) Vascular plants of Nebraska
(Version 1 -July 2003). Nebraska Game and Parks Commission. Lincoln, NE 57
pp. List was updated in 2011 and forwarded via email from G. Steinaur to
Nicole Kirchner in 2011.
NH
NEIW13_NH
New England Interstate Water Pollution Control Commission (NEIWPCC).
(2013) Northeast Regional Floristic Quality Assessment. Current URL (27
August 2019): https://neiwpcc.org/our-programs/wetlands-aquatic-
species/nebawwg/nqa/. Individual State CoC lists: New Hampshire.
NJ
WALZ17
Walz, K. S., Kelly, L. and Anderson, K. (2017) Floristic Quality Assessment
Index for Vascular Plants of New Jersey: Coefficient of Conservancy (CoC)
Values for Species and Genera. New Jersey Department of Environmental
Protection, New Jersey Forest Service, Office of Natural Lands Management,
Trenton, NJ, 08625. Submitted to United States Environmental Protection
Agency, Region 2, for State Wetlands Protection Development Grant, Section
104(B)(3); CFDA No. 66.461, CD97225809.
NJ
WALZ19
Walz, Kathleen S., Linda Kelly, Karl Anderson, Keith Bowman, Barbara
Andreas, Richard Andrus, Scott Schuette, William Schumacher, Sean
Robinson, Terry O'Brien, Eric Karlin and Jason Hafstad (2019). Universal
Floristic Quality Assessment Index for Vascular Plants and Mosses of New
Jersey: Coefficient of Conservancy (CoC) Values for Species and Genera
(Updated November 2019). New Jersey Department of Environmental
Protection, New Jersey Forest Service, Office of Natural Lands Management,
Trenton, NJ, 08625. Submitted to United States Environmental Protection
Agency, Region 2, for State Wetlands Protection Development Grant, Section
104(B)(3); CFDA No. 66.461, CD97225809.
NJ
KELL13
Kelly, L., Anderson, K. & Walz, K.S. (2013) New Jersey floristic quality
assessment: coefficients of conservatism for vascular taxa
NM
EPA19_NM
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019). Project to
assign C-values to western states for use in the USEPA National Wetland
Condition Assessment (NWCA): C-values for taxon-state pairs observed in
AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA Surveys.
Funded by USEPA Office of Wetlands, Oceans, and Watersheds to the Great
Lakes Environmental Center, Traverse City, Ml. EP-C-16-008: Task Order #08.
Unpublished Report and Excel File (NWCA_C_Values Western States_ll-6-
2018_Draft) Submitted to USEPA.
NM
UIUC19
Jessica L. Stern, Brook D. Herman, Jeffrey W. Matthews (2021). Coefficients
of conservatism for the flora of the middle Rio Grande floodplain. The
Southwestern Naturalist, 65(2), 141-151.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
NV
EPA19_NV
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019). Project to
assign C-values to western states for use in the USEPA National Wetland
Condition Assessment (NWCA): C-values for taxon-state pairs observed in
AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA Surveys.
Funded by USEPA Office of Wetlands, Oceans, and Watersheds to the Great
Lakes Environmental Center, Traverse City, Ml. EP-C-16-008: Task Order #08.
Unpublished Report and Excel File (NWCA_C_Values Western States_ll-6-
2018_Draft) Submitted to USEPA.
NY
NEIW13_NY
New England Interstate Water Pollution Control Commission (NEIWPCC)
(2013) Northeast Regional Floristic Quality Assessment. Current URL (27
August 2019): https://neiwpcc.org/our-programs/wetlands-aquatic-
species/nebawwg/nqa/. Individual State CoC lists: New York.
OH
ANDR04
Andreas, B.K., J.J. Mack, and J.S. McCormac (2004) Floristic quality
assessment index (FQAI) for vascular plants and mosses for the State of
Ohio. Ohio Environmental Protection Agency, Division of Surface Water,
Wetland Ecology Group, Columbus, OH. 219 pp.
OK
EWIN12
Ewing, A.K., and Hoagland, B. (2012) Development of floristic quality index
approaches for wetland plant communities in Oklahoma. USEPA Final
Report, FY201,104(b)(3), CD-00F074, Project 2.
OR
EPA19_OR
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019). Project to
assign C-values to western states for use in the USEPA National Wetland
Condition Assessment (NWCA): C-values for taxon-state pairs observed in
AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA Surveys.
Funded by USEPA Office of Wetlands, Oceans, and Watersheds to the Great
Lakes Environmental Center, Traverse City, Ml. EP-C-16-008: Task Order #08.
Unpublished Report and Excel File (NWCA_C_Values Western States_ll-6-
2018_Draft) Submitted to USEPA.
OR
MAGE01
Magee, T.K. and Bollman, M.A. (2013, unpublished). C-values for ~500
Streamside plant species in eastern Oregon.
Rl
NEIW13_RI
New England Interstate Water Pollution Control Commission (NEIWPCC).
(2013) Northeast Regional Floristic Quality Assessment. Current URL (27
August 2019): https://neiwpcc.org/our-programs/wetlands-aquatic-
species/nebawwg/nqa/. Individual State CoC lists: Rhode Island.
SD
TNGP01
The Northern Great Plains Floristic Quality Assessment Panel, (2001)
Coefficients of conservatism for the vascular flora of the Dakotas and
adjacent grasslands: US Geological Survey, Biological Resources Division,
Information and Technology Report USGS/ BRD/ITR—2001-0001, 32 p.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
112
-------
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
TN
TN_CC
Compiled from:l) Willis, K. and Estes, L unpub. 2013. Floristic Quality
Assessment for Tennesse Vascular Plants, 2) Gianopulos, K. (2014)
Coefficient of Conservatism Database Development for Wetland Plants
Occurring in the Southeast United States: Summary Document. North
Carolina Dept. of Environment and Natural Resources, Division of Water
Resources. See: USEPA (2016) National Wetland Condition Assessment: 2011
Technical Report. EPA-843-R-15-006. Section 5.9 Species Traits - Coefficients
of Conservatism.
TX
EPA19_TX
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019). Project to
assign C-values to western states for use in the USEPA National Wetland
Condition Assessment (NWCA): C-values for taxon-state pairs observed in
AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA Surveys.
Funded by USEPA Office of Wetlands, Oceans, and Watersheds to the Great
Lakes Environmental Center, Traverse City, Ml. EP-C-16-008: Task Order #08.
Unpublished Report and Excel File (NWCA_C_Values Western States_ll-6-
2018_Draft) Submitted to USEPA.
TX
Reemtzl9
Reemts, C. M., and J. A. Eidson (2019). Choosing Plant Diversity Metrics: A
Tallgrass Prairie Case Study. Ecological Restoration 37:233-245.
UT
EPA19_UT
Fennessy, M. S., & Great Lakes Environmental Center, Inc (2019). Project to
assign C-values to western states for use in the USEPA National Wetland
Condition Assessment (NWCA): C-values for taxon-state pairs observed in
AZ, CA, ID, NV, NM, OR, TX, UT during the 2011 and 2016 NWCA Surveys.
Funded by USEPA Office of Wetlands, Oceans, and Watersheds to the Great
Lakes Environmental Center, Traverse City, Ml. EP-C-16-008: Task Order #08.
Unpublished Report and Excel File (NWCA_C_Values Western States_ll-6-
2018_Draft) Submitted to USEPA.
VA
VDEP05
Virginia Department of Environmental Quality (2005) Determining
coefficient of conservatism values (C-Values) for vascular plants frequently
encountered in tidal and nontidal wetlands in Virginia. Report prepared for
US Environmental Protection Agency-Region III. Wetlands Program
Development Grant #CD983380-01
VT
NEIW13_VT
New England Interstate Water Pollution Control Commission (NEIWPCC).
(2013) Northeast Regional Floristic Quality Assessment. Current URL (27
August 2019): https://neiwpcc.org/our-programs/wetlands-aquatic-
species/nebawwg/nqa/. Individual State CoC lists: Vermont.
WA
ROCC13
Roccio, F.J. & Crawford, R.C. (2013) Floristic quality assessment for
Washington vegetation. Washington Natural Heritage Program Washington
Department of Natural Resources. Natural Heritage Report 2013-03. USEPA
Wetland Program Development Grant Assistance Agreements: 1) CD-
00J26301 and CD-00J49101
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
113
-------
State or Region
Source
Abbreviation
Coefficient of Conservatism Lists included in the Compilation of Existing C-
values
Wl
BERN03
Bernthal, T.W. (2003) Development of a Floristic Quality Assessment
methodology for Wisconsin. Report to the USEPA (Region V). Wisconsin
Department of Natural Resources: Bureau of Fisheries Management and
Habitat Protection. USEPA Wetland Grant # CD975115-01-0
Wl
CHUN17
Chung-Gibson, M., Bernthal T., Doyle K., Wetter, M., Haber, E. (2017).
Wisconsin Department of Natural Resources, Water Quality Bureau. From
WDNR_FQA_Calculator_vl.5.17. Nomenclature from Wisconsin State
Herbarium, University of Wisconsin-Madison (2016). COFC values from
Bernthal, TW. Development of a Floristic Quality Assessment Methodology
for Wisconsin. Wisconsin Department of Natural Resources, 2003. Note that
regions differ only in Wetland Indicator Status.
Wl
PARK14
Parker, E.C., Curran, M., Waechter, Z.S. and Grosskopf, E.A. (2014) Wisconsin
FQA (Floristic Quality Assessment) Databases for Midwest and Northcentral-
Northeast Regions for Universal FQA Calculator.
WV
RENT06
Rentch, J.S.& Anderson, J.T. (2006) A floristic quality index for West Virginia
wetland and riparian plant communities. Division of Forestry and Natural
Resources, West Virginia University. US Department of Agriculture CREES,
Award No. 2004-38874-02133.
WV
WVHP15
West Virginia Natural Heritage Program (2015) Coefficients of Conservatism
for the Vascular Flora of West Virginia. Wildlife Diversity Unit, West Virginia
Division of Natural Resources, Elkins, West Virginia, USA.
WY
WASH15
Washkoviak L, Heidel, B, and Jones, G (2017). Floristic Quality Assessment
for Wyoming Flora: Developing Coefficients of Conservatism. Prepared for
the US Army Corps of Engineers. The Wyoming Natural Diversity Database,
Laramie, Wyoming. 13 pp. plus appendices.
Mid-Atlantic
(Mid_Atl)
Region
CHAM12
Chamberlin J, Ingram H (2012) Developing coefficients of conservatism to
advance floristic quality assessment in the Mid-Atlantic region.
Northeast
(NEngl) Region
FABE18
Faber-Langendoen, D. (2018) Northeast Regional Floristic Quality
Assessment Tools for Wetland Assessments. NatureServe, Arlington VA
Southeast
(SEast) Region
GIAN14
Gianopulos, K. (2014) Coefficient of Conservatism Database Development for
Wetland Plants Occurring in the Southeast United States: Summary
Document. North Carolina Dept. of Environment and Natural Resources,
Division of Water Resources.
Southeast
(SEast) Region
SE CoC(2023)
NC Division of Water Resources. 2018-2024. North Carolina Wetlands
Information, https://www.ncwetlands.org. Published by the North Carolina
Division of Water Resources, Water Sciences Section
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
114
-------
Chapter 8: Vegetation Analyses and Candidate Metric Evaluation
Prerequisite to Multimetric Index Development
8.1 Overview
The analysis of the vegetatation data for NWCA 2021
used the same Vegetation Multimetric Indexes (VMMI)
developed for the 2016 survey to assess biological
condition. This chapter describes the process to evaluate
candidate metrics that were a prerequite step to
developing the VMMIs for the broad wetland groups
used in NWCA 2016.
Data from both the 2011 and 2016 NWCA surveys
(Figure 8-1) was used to develop the VMMIs. For sites
that had repeat sampling events, the data from the
Index Visit to that site were used for developing the
disturbance gradient (Chapter 6:) and for developing the
VMMIs. 1,987; unique sites were used in setting the
disturbance gradient (Table 6-1); however, at two of
these sites, vegetation data were not collected.
Consequently, the Index Visit data from 1985 NWCA
sites where vegetation was sampled in 2011 or 2016
(Table 8-1) were used in calculating and evaluating
candidate vegetation metrics and developing four
Wetland Group VMMIs (discussed further in Chapter 9:).
Several initial analysis steps were needed to support
development of the NWCA VMMIs:
Step 1: Definition of anthropogenic disturbance gradients by identifying least- and most-disturbed
sites (Section 8.2 and Chapter 6:).
Step 2: Consideration of sample sizes and variability in species composition across regions and
wetland types to determine potential scales (e.g., national, wetland type, ecoregion) for metric
evaluation and VMMI development (Section 8.3).
Step 3: Calculation (Section 8.4) of candidate vegetation metrics.
Step 4: Evaluation of candidate vegetation metrics (Section 8.5) for use in VMMI development.
In addition to the Index Visit data, where unique sites also had a sampling revisit (Visit 2) during the same
field season, these revisit data were compared to data of the Index Visit to calculate signal:noise (S:N)
ratios, which were used in aspects of metric (Section 8.5.2) and VMMI (Section 9.3.2) screening.
Analyses were completed with R Statistical Software, ver. 3.6.1 (R Core Team 2019), except detrended
correspondence analysis for which PC-ORD, ver. 7.8 (McCune and Mefford 2018) was used.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 8-1. Numbers of unique NWCA 2011 and 2016 sampled sites. NWCA_REF (Disturbance): L = Least, I =
Intermediate, M = Most, ? = Undetermined, Revisit = site sampled twice in same field season.
n Numbers of Unique Sites by Type
ALL SITES
Total
L
1
M
?
Revisit
Calibration
Validation
1985
439
1061
474
11
104
1587
398
NWCARPT UNIT 6
Inland Coastal Plains (ICP)
Eastern Mountains and Upper Midwest (EMU)
Plains (PLN)
Arid West (ARW)
I Western Valleys & Mountains (WVM)
¦ Tidal (TDL)
o 2011 Probability Sites
A 2011 Handpicked Sites
• 2016 Probability Sites
A 2016 Handpicked Sites
Figure 8-1. Distribution of probability and hand-picked sites sampled in the 2011 and 2016 NWCA surveys within
Six Reporting Units (RPT_UNIT_6). TDL = coastal areas where tidally-influenced estuarine wetlands occur. Inland
wetlands are mapped within five geographic regions.
8.2 Anthropogenic Disturbance
Both the evaluation of candidate metrics for utility in reflecting ecological condition and the development
of VMM Is require least-- and most-disturbed sites to anchor the ends of an anthropogenic disturbance
gradient (USEPA 2016a, Magee et al. 2019a). In addition, least-disturbed sites are used in setting
thresholds for good, fair, and poor condition based on VMMI values (Magee et al. 2019a, Herlihy et al.
2019).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
116
-------
The multi-step process for screening and assigning least-disturbed, intermediate-disturbed, or most-
disturbed status to NWCA sites is detailed in Chapter 6: and summarized in Appendix A: Illustrative
Guide to Assigning Disturbance Class in Six Steps. In brief, a stepwise process was used in which sites
were first screened for abiotic disturbance using physical indices (Section 6.3), then by chemical indices
(Section 6.4) to assign abiotic disturbance classes. Least-disturbed sites passing the physical and chemical
screens (Section 6.5), were further screened with a biological metric (XRCOV_AC, (Section 6.6), the
relative percent cover of nonnative (alien and cryptogenic, Table 7-5) plants. The final set of least-,
intermediate-, and most-disturbed sites (REF_NWCA) was used for evaluation of vegetation candidate
metrics and for VMMI development based on the Index Visit data from 1985 unique NWCA sites where
vegetation was sampled in 2011 or 2016 (Table 8-1).
8.3 Considering Regional and Wetland Type Differences
To account for physical and biotic diversity across the national scale, finer scales are often needed to
facilitate development of the most effective MMIs (Stoddard et al. 2008, USEPA 2006, Herlihy et al.
2019). Plant species composition in wetlands varies widely across the conterminous United States, both
with environmental conditions and wetland type (Herlihy et al. 2019, USEPA 2016a, Magee et al. 2019a).
We evaluated a series of potential subpopulation groups (Table 5-1) in an effort to minimize natural
variation, while maintaining sample sizes sufficient to inform candidate metric evaluation and VMMI
development. To identify scales relevant for VMMI development based on the plant data from NWCA
2011 and NWCA 2016 sampled sites, we examined the following groupings listed from finer to coarser
scale:
• RPT_UNIT12 (Table 8-2, Figure 8-2): 12 subpopulations based on combining region (RPT_UNIT_6)
and wetland group (WETCLS_GRP)
• RPT_UNIT_6 (Table 8-3): six subpopulations including tidally-infIueneed estuarine wetlands in
coastal areas and inland wetlands in 5 aggregated ecoregions
• WETCLS_GRP (Table 8-4): four subpopulations describing broad Wetland Groups
Table 8-2 through Table 8-4 include, for each subpopulation: 1) the total number of unique sampled sites;
2) the numbers of sites identified as "least disturbed", "intermediate disturbed", and "most disturbed"; 3)
the number of revisit sites (sites sampled twice during the same sampling season to quantify within-year
sampling variability); and 4) the number of calibration and validation sites used in analyses. Ordination
analysis of the plant species data was used to evaluate how species composition (presence and
abundance) varied in relation to these broad ecoregional and wetland group subpopulations (Figure 8-3).
Table 8-2. Numbers of unique NWCA 2011 and NWCA 2016 sampled sites by RPT_UNIT12 (RPT_UNIT_6 x
WETCLS_GRP). RPT_UNIT_6 is defined in Figure 6-2 and Table 6-3. WETCLS_GRP is defined in Table 6-4.
REF_NWCA (Disturbance): L = Least, I = Intermediate, M = Most, ? = undetermined. Revisit = site sampled twice in
same field season.
RPT_UNIT12
(RPT_UNIT_6x
WETCLS_GRP)
RPT_GRP_12*
n Numbers of Sites by Type
Total
L
1
M
?
Revisit
Calibration
Validation
TDL-H
ALL-EH
374
134
158
81
1
18
298
76
TDL-W
ALL-EW
87
15
43
27
2
4
70
17
ICP-H
CPL-PRLH
104
21
48
34
1
3
83
21
ICP-W
CPL-PRLW
307
65
168
71
3
11
247
60
EMU-H
EMU-PRLH
116
29
61
26
0
11
90
26
117
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
EMU-W
EMU-PRLW
234
72
115
46
1
17
181
53
PLISI-H
IPL-PRLH
210
19
124
65
2
15
169
41
PLN-W
IPL-PRLW
141
34
84
23
0
4
121
20
ARW-H
XER-PRLH
109
7
70
32
0
2
86
23
ARW-W
XER-PRLW
59
3
43
13
0
3
40
19
WVM-H
WMT-PRLH
113
20
63
30
0
8
94
19
WVM-W
WMT-PRLW
131
20
84
26
1
8
108
23
*Note: membership of sites in subpopulations of RPT_UNIT12 and of RPT_GRP_12 is the same. RPT_UNIT12 codes
were created to allow matching with codes in RPT_UNIT_6 and WETCLS_GRP.
V?
RPT UNIT 6
\ o a t f o • • _ L .
.v
| Inland Coastal Plains (ICP) \ •
Eastern Mountains and Upper Midwest (EMU) *
Plains (PLN) V
¦ Arid West (ARW)
H Western Valleys & Mountains (WVM)
¦ Tidal (TDL)
WETCLS_GRP
A Estuarine Herbaceous (EH)
a Estuarine Woody (EW)
° Palustrine, Riverine, and Lacustrine Herbaceous (PRLH)
• Palustrine, Riverine, and Lacustrine Woody (PRLW)
Figure 8-2. Six Reporting Units and four Wetland Groups: TDL = coastal areas where tidally-influenced estuarine
wetlands occur. Inland wetlands are mapped within 5 NWCA Aggregated Ecoregions.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
118
-------
Table 8-3. Numbers of unique NWCA 2011 and 2016 sampled sites by six reporting units (RPT_UNIT_6).
REF_NWCA (Disturbance): L = Least, I = Intermediate, M = Most, ? = undetermined. Revisit = site sampled twice in
same field season. Tidal (TDL) = tidally-influenced estuarine wetlands occurring in near coastal areas. The other five
groups represent inland wetlands within five ecoregional areas. See Table 8-4 for description of include wetland
types.
RPT_UNIT_6
n Numbers of Sites by Type
Total
L
1
M
?
Revisit
Calibration
Validation
TDL
Tidal
461
149
201
108
3
22
368
93
ICP
Inland Coastal Plains
411
86
216
105
4
14
330
81
EMU
Eastern Mtns & Upper
Midwest
350
101
176
72
1
28
271
79
PLN
Interior Plains
351
53
208
88
2
19
290
61
ARW
Arid West
168
10
113
45
0
5
126
42
WVM
Western Valley & Mountains
244
40
147
56
1
16
202
42
Table 8-4. Numbers of unique NWCA 2011 and 2016 sampled sites by Wetland Groups (WETCLS_GRP). REF_NWCA
(Disturbance): L = Least, I = Intermediate, M = Most, ? = undetermined. Revisit = site sampled twice in same field
season. EH and EW are tidally-influenced estuarine wetlands. PRLH and PRLW are inland wetlands.
WETCLS_GRP1 (Wetland Groups)
n Numbers of Sites by Type
Total
L
1
M
?
Revisit
Calibration
Validation
EH
Estuarine Herbaceous
374
134
158
81
1
18
298
76
EW
Estuarine Woody
87
15
43
27
2
4
70
17
PRLH
Palustrine, Riverine or
Lacustrine Herbaceous
654
96
366
187
3
39
522
130
PRLW
Palustrine, Riverine or
Lacustrine Woody
872
194
494
179
5
43
697
175
1Wetland types included in each WETCLS_GRP category listed above are defined below
WETCLS GRP
Description of wetland types included
NWCA
Wetland
Type
USFWS
Status &
Trends Code
Estuarine
EH
Estuarine intertidal (E) emergent (H = herbaceous)
EH
E2EM
EW
Estuarine intertidal (E) forested and shrub (W=
woody)
EW
E2SS
Inland
PRLH
Emergent wetlands (EM) in palustrine, shallow
riverine, or shallow lacustrine littoral settings (PRL)
PRL-EM
PEM
Farmed wetlands (f) in palustrine, shallow riverine,
or shallow lacustrine littoral settings (PRL); only
subset previously farmed, but not currently in crop
production
PRL-f
Pf
Open-water ponds and aquatic bed wetlands
PRL-UBAB2
PUBPAB2
PRLW
Shrub-dominated wetlands (SS) in palustrine,
shallow riverine, or shallow lacustrine littoral
settings (PRL)
PRL-SS
PSS
Forested wetlands in palustrine (FO), shallow
riverine, or shallow lacustrine littoral settings (PRL)
PRL-FO
PFO
2PUBPAB covered S&T Wetland Categories: PAB (Palustrine Aquatic Bed), PUBn (Palustrine Unconsolidated Bottom, natural),
PUBa (aquaculture), PUBf (agriculture use), PUBi (industrial), PUBu (PBU urban).
119
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Ideally, for VMMI development, each subpopulation or analysis group would have at least 100 total sites,
with 30 of these meeting least-disturbed criteria. Not all of these potential analysis groups (Table 8-2
through Table 8-4) had the recommended number of total sites or of least-disturbed sites for VMMI
development. For example, among the RPT_UNIT12 categories (Table 8-2) the tidally-influenced Estuarine
Woody wetlands (TDL-W) and the inland Arid West woody wetlands (ARW-PRLW) each had fewer than
100 sites, and several region-Wetland Group combinations had fewer than 30 least-disturbed sites. At the
coarser scale of RPT_UNIT_6 subpopulations, only 10 least-disturbed sites were available for the Arid
West (ARW) (Table 8-3). In the WETCLS_GRP classification (Table 8-4), there were only 87 Estuarine
Woody (EW) wetland sites and only 15 of these were least-disturbed.
Ordination using detrended correspondence analysis (DCA) (McCune and Mefford 2018) illustrated how
species composition, based on species identity and abundance (estimated as percent cover), at the site-
level varied in relation to broad wetland type and ecoregional subpopulations. The DCA was based on the
percent cover of 4,798 observed taxa (native and nonnative) observed in one or more of the sampled
sites and was run with down-weighting of uncommon taxa and axis rescaling (segments = 30).
Eigenvalues for axes 1 and 2 were 0.949 and 0.803, respectively, with a Monte Carlo randomization test
(999 permutations) having p = 0.0001 for both axes. Total variance in the species data was 113.3. The
ordination (Figure 8-3) was plotted using raw site scores and unrotated axes (McCune and Mefford 2018),
with sites coded to represent the 12-Ecoregion x Wetland Group (RPT_UNIT12) subpopulations.
1200 ¦
1000 ¦
800 ¦
(N
U)
X
<
600 ¦
400 ¦
200 ¦
*
. * i£.
* ^
RPTJJNIT12
+ ARW-H
+ ARW-W
EMU-H
a EMU-W
v ICP-H
t ICP-W
PLN-H
~ PLN-W
oTDL-H
• TDL-W
~ WVM-H
¦ WVM-W
0 ¦
200
400
600
800
Axis 1
1000
1200
1400
1600
Figure 8-3. Detrended correspondence analysis for NWCA 2011 and 2016 sampled sites. Sites are color- and
symbol-coded by RPT_UNIT12. Blue and TDL = Tidally-influenced, estuarine wetland sites. Other codes and colors =
Inland wetland sites by geographic region. Open symbols = herbaceous (H) wetlands. Filled symbols = woody (W)
wetlands. Note: Among the unique 1985 sampled sites, 208 were resampled sites (sampled in 2011 and 2016),
(Section 6.1), and for these resampled sites the data from 2011 visit were used in this DCA.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
DCA axis gradient length reflects the standard deviations (SD) in species composition along an axis; and
sites with scores that differ by more than 4 SD are expected to have no species in common (McCune and
Grace 2002, Jongman et al. 1995). Gradient length for Axis 1 was 14.9 and for Axis 2 was 11.8. This means
that from one edge of Axis 1 to the other (i.e., moving from left to right across the ordination), there are
3.7 complete turnovers in species composition. Similarly, for Axis 2 (i.e., moving from top to bottom of
the ordination), there are nearly 3 turnovers in species composition. This level of beta diversity is not
surprising given the geographic scope of the study area (conterminous US) and the diversity of wetland
plant communities that are represented in each 12-Ecoregion x Wetland Group subpopulation.
The ordination plot (Figure 8-3) shows distinct to intergrading groups of sites associated with wetland
subpopulations along gradients described by the axes. Tidal (EH, EW) vs. inland (PRLH, PRLW) wetlands
separate distinctly along Axis 1. Inland herbaceous and woody wetlands tend to separate within
ecoregional groups along both Axes 1 and 2. Axis 2 appears to be related to longitude, with sites from the
western half of the US (e.g., WVM, ARW, PLNS) tendingto occur in the top half of the ordination and
those from the eastern half (EMU, ICP) occurring in the bottom half. Within the ecoregional groups the
woody sites separate more distinctly than the herbaceous sites, and woody wetlands tend to be
distributed along the outer edges/portions of the ordination by their ecoregional groups. The woody
(PRLW) wetlands in the PLNS tend to intermix at the interface between EMU-PRLW and ICP-PRLW Some
Inland Herbaceous wetlands (PRLH in the WMV, ARW, and PLNS), tend to intermix in the center and
upper right of the ordination and to intergrade more among regions than do the woody wetlands.
Intermixing of these herbaceous Wetland Groups may be related in part to the presence of widespread
native species and to nonnative species with wide ecological amplitude (Magee et al 2019b).
The ordination of these 12-Ecoregion x Wetland Type (RPT_UNIT12) subpopulations was useful in
describing variation in wetland vegetation at a continental scale, with wetland type (WETCLS_GRP)
appearing to be primary and ecoregion (RPT_UNIT_6) to be secondary drivers of species composition.
Given these patterns and the available sample sizes for least-disturbed sites in the various classifications
(Table 8-1 through Table 8-4), we evaluated metrics (Section 8.5) and developed candidate VMMIs
(Section 9.3) at the national scale, and for subpopulations in WETCLS_GRP and in RPT_UNIT_6:
• National scale - all sampled wetlands (Table 8-1)
• Five subpopulations based on RPT_UNIT_6 groups (Figure 8-2 and Table 8-3):
o TDL - tidally-influenced estuarine wetlands in coastal areas
o Inland wetlands in Four NWCA Aggregated Ecoregions
¦ ICP - Inland Coastal Plains
¦ EMU - Eastern Mountains and Upper Midwest
¦ PLNS_ARW- Plains (PLN) and Arid West (ARW); note the PLN and ARW groups
were combined because there were few least-disturbed sites in ARW
¦ WVM - Western Mountains and Valleys
• Four Wetland Group subpopulations (WETCLS_GRP) (Table 8-4)
Candidate VMMIs for all these groups were developed and evaluate to identify which might have the
most robust performance.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
121
-------
8.4 Calculating Candidate Metrics
Validated vegetation data (see Sections 7.4 and
7.5), along with species trait information, (see
Sections 7.6 through 7.9) were used to calculate
numerous candidate metrics representing several
major Metric Groups (Table 8-5). These
ecologically important Metric Groups and their
component metrics are commonly recognized as
potential indicators of condition or stress (USEPA
2016a, Magee et al 2019).
The Metric Groups listed in Table 8-5 are
comprised of a variety of broad metric types, and
for each metric type, several-to-many specific
candidate metrics with potential relationships to
ecological condition or stress were calculated.
NWCA candidate vegetation metrics included
descriptors that were likely to have broad
applicability across regions and wetland types, as
well as metrics expected to have more restricted
utility for specific broad wetland groups. Section
8.8, Appendix E lists: 1) the name and a short
description of each metric, 2) how each metric was calculated, 3) the field data and species trait groups
on which each metric is based, and 4) whether the metric is used primarily to describe ecological
condition or stress in the NWCA.
The metric information specified in Section 8.8, Appendix E was used in updating R code to calculate 556
candidate vegetation metrics for each sampled site. The original, accuracy-tested, R code that was
developed for metric calculation for the 2011 survey (USEPA 2016a) was updated, here, for the joint
analysis of the 2011 and 2016 data. The calculated metrics can be found on the NWCA website
(.nwca_2016_ veg_metrics. csv).
Most of the metric types described in Table 8-5 include versions of metrics that incorporate all species,
only native species, or only nonnative species. Vegetation metrics based on all species or on only native
species were considered as potential descriptors of wetland condition (n = 426). Metrics based on only
nonnative species (alien and cryptogenic species, see Section 7.8) (n = 130) were viewed as indicators of
wetland stress (USEPA 2016a). Only the former group of metrics was considered in VMMI development.
The 426 candidate condition metrics were used in developing candidate VMMIs (see Chapter 9:). In
previous work, the Nonnative Plant Indicator (NNPI) was developed based on data from the 2011 NWCA
(Magee et al. 2019a). Here, the NNPI was applied in analysis of the combined 2011 and 2016 data
(Chapter 10:). The NNPI uses exceedance values for three nonnative plant metrics to assign categorical
classes (good, fair, poor, and very poor) to describe wetland condition in relation to impact from
nonnative plants.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
122
-------
Table 8-5. Metric Groups and component Metric Types for characterizing vegetation condition.
Metric Groups
Major Metric Types for each Metric Group
Taxa Composition3
Richness, diversity, frequency, cover, importance of vascular plant species, genera,
families, etc.
Floristic Quality3
Mean Coefficient of Conservatism, Floristic Quality Assessment Index (versions
based on species presence or frequency and cover-weighted versions)
Tolerance and Sensitivity
to Disturbance
Richness and abundance of sensitive, insensitive, tolerant, highly tolerant species
Hydrophytic Status3
Richness and abundance by Wetland Indicator Status; Wetland Indices
Life History3
Richness and abundance by growth-habit type, duration/longevity category,
vascular plant category (e.g., ferns, dicots, etc.)
Vegetation Structure
Frequency, cover, importance, diversity, by structural (height) vegetation groups
Nonvascular
Frequency, cover, importance for ground or arboreal bryophytes or lichens, algae
Ground Surface Attributes
Frequency, cover, importance, depth of water, litter, bare ground
Woody Debris and Snags
Frequency, cover, importance for woody debris, counts for snags
Trees3
Richness, counts, or frequency, cover or importance by height or diameter classes
individual metrics in a group often included versions based on all species or native species only. Note: All
importance metrics combine frequency and cover.
Only a small number of the calculated metrics were ultimately incorporated in NWCA vegetation indices
(VMMIs, Chapter 9:) or (NNPI, Chapter 10:, Magee et al. 2019b). However, many of the other metrics are
expected to be useful in describing other characteristics of wetlands or for addressing ecological
questions related to diversity, structure, functional traits, or relationships to environmental conditions or
ecological processes. For example, the nonnative plant metrics (n =130) are likely to inform questions
related to the impacts of nonnative plants, which can 1) reflect condition of the vegetation, 2) be
indicators of anthropogenic disturbance, or 3) behave as direct stressors to vegetation and ecosystem
properties (e.g., Kuebbing et al. 2015, Magee et al. 2008, 2010, 2019b, Pysek et al. 2020, Riccardi et al.
2020, Ruaro et al 2020, Simberloff 2011).
8.5 Evaluating Candidate Vegetation Metrics
Data from all 1,985 unique 2011 and 2016 sampled sites were used to evaluate 426 individual NWCA
candidate vegetation metrics of condition for their potential utility in development of candidate VMMI(s).
The NWCA metric screening approach was adapted and expanded for wetlands (Magee et al. 2019a) from
metric evaluation methods used in other NARS (e.g., Stoddard et al. 2008, Pont et al. 2009, VanSickle
2010). Most of the wetland vegetation metrics were strongly non-normal (Magee et al. 2019a, USEPA
2016a); consequently, nonparametric statistical (e.g., Kruskal-Wallis test) approaches were used in the
screening analyses where appropriate. Specific criteria for range, repeatability, responsiveness, and
redundancy were defined. R code was written to implement these screening tests.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
8.5.1 Range Tests
Metrics with limited range, too many zero values, or highly skewed distributions have been shown to
generally be poor indicators of ecological condition. Thus, sufficient range in values to permit signal
detection is important. We used two tests to define sufficient (PASS), marginal (PASS-), and insufficient
(FAIL) range for metric values.
• Test 1 - Identifies metrics with large proportion of 0 values or highly skewed distributions:
o If the 75th percentile = 0, i.e., more than 75% of values are 0, then FAIL
o If the 75th percentile = the minimum OR the 25th percentile = max (indicating 75% of
values identical), then FAIL (ensures that a majority of values are not the same as the
minimum or maximum to help eliminate variables that are highly skewed and mostly a
single non-zero value)
o If the median = 0, then PASS-
• Test 2 - Identifies metrics with very narrow ranges
o If the metric is a percent variable and (max-25th percentile) < 15%, then FAIL
o If the metric is not a percent variable and (max-25th) < (max/3), then FAIL
If either Test 1 or 2 resulted in a FAIL, the final assignment for the metric was FAIL. If the first two screens
in Test 1 resulted in a PASS, but the third screen a PASS-, the result was PASS-. To pass the range screen,
each metric had to receive a PASS or PASS-.
8.5.2 Repeatability (S:N)
Useful metrics tend to have high repeatability, that is, the among-site variability will be greater than
within-year sampling variability based on repeat sampling during the same field season at a subset of sites
(see Table 8-1 through Table 8-4, revisit sites). To quantify repeatability, NARS uses Signal-to-NoiseRatio
(S:N), that is, the ratio of variance associated with a sampling site (signal) to the variance associated with
repeated visits to the same site (noise) (Kaufmann et al. 1999). All sites are included in the signal, whereas
only revisit sites contribute to the noise component. Metrics with high S:N are more likely to show
consistent responses to human-caused disturbance, and S:N values < 1 indicate that sampling a site twice
yields as much or more metric variability as sampling two different sites (Stoddard et al. 2008).
In the NWCA, we set an initial criterion of S:N > 4 (Magee et al. 2019a). In practice, however, the
observed S:N values for the vegetation metrics were much higher, so we ultimately set the metric
retention criterion to S:N > 10, or > 5 if metric type was as yet unrepresented in the suite of metrics
passing all selection criteria. For the NWCA, S:N for individual metrics was calculated using the R package
"Ime4" (version 1.1-7, Bates et al. 2014). Each metric was used as a response variable with SITEJD (a site
identifier) as the main factor in a random effects model. Then the variance components from the
resulting model were used to calculate S:N.
Note, that among the analysis groups for which metric screening was conducted (Section 8.5.5), two
subpopulations had < 5 revisit sites (ARW, EW). For these, two groups S:N values were given little
consideration compared to other screening criteria.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
8.5.3 Responsiveness
The most fundamental test of the efficacy of a candidate metric is its capacity to discriminate degraded
from relatively undisturbed ecosystems. Responsive candidate metrics effectively distinguish least-
disturbed from most-disturbed sites (Stoddard et al. 2008). In the NWCA, the ability to differentiate least-
from most-disturbed sites was evaluated based on p-values and Chi-squared values from a Kruskal-Wallis
test (large sample approximation). The assessment of the discriminatory capability of individual metrics
was also supported by ranking the separation of least- and most-disturbed sites based on box plot
comparisons, where the degree of overlap of medians and interquartile ranges (IQRs) between least- and
most-disturbed sites provides a signal of the metric responsiveness (Klemm et al. 2002).
We developed R code to automate a process to simulate comparison of box plots for least and most
disturbed sites, for each vegetation metric, and to rank the separation levels. Using the approach
developed by Barbour et al. (1996) and outlined in Klemm et al. (2002), the medians and IQRs of the least
and most disturbed sites were compared, and metrics were scored as follows:
• Score of 0 (lowest discriminatory power) - Complete overlap of each group's IQRs with the
median of the other group
• Score of 1 - Only one median was overlapping with the IQRs of the other group
• Score of 2 - Neither median overlapped with the IQR of the other group, but the IQRs
overlapped
• Score of 3 (highest discriminatory power) - IQRs did not overlap
Metric responsiveness was evaluated using three acceptance thresholds:
• Kruskal-Wallis p < 0.05
• Chi-square value from Kruskal-Wallis test >10, or >5 if metric type was as yet unrepresented in
the suite of metrics passing all selection criteria
• Box plot separation score > 0
o A zero-value box plot did not disqualify if the metric passed the other screens and was
not represented in the suite of metrics passing all other selection criteria
o Higher box plot separation scores received greater preference (3 > 2 > 1) in selecting
among related metrics
Among metrics passing the responsiveness screen, the Kruskal-Wallis p-values were often much lower
and Chi-square values were often much higher than acceptance thresholds. In some cases where other
screening criteria were high, a metric with Chi-square < 5 might be retained.
8.5.4 Redundancy
Step 1- During metric screening, a subset of metrics that passed the range, repeatability, and
responsiveness tests, but which conveyed information similar to other metrics, were dropped. Dropped
metrics typically included those that were very similar (e.g., absolute versus relative cover for trait-based
metrics) or individual metrics that were also represented as a component of another metric. In such
cases, the metric that was considered most ecologically meaningful, performed best on screening tests,
or was easiest to collect or calculate was selected.
Step2- Additional redundancy screening was handled during the process of VMMI development. It is
generally agreed that metrics included in a MMI should not be strongly correlated, and r < 0.75 is often a
cut off point for correlation among metrics included in the same MMI (e.g., Stoddard et al. 2008, Pont et
125
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
al 2009, Van Sickle 2010). Candidate VMMIs were screened to ensure that correlations among their
component metrics were less than this threshold. If this threshold was exceeded the candidate VMMI
was disqualified (see Section 9.3).
8.5.5 Application of Metric Screening Criteria
Screening criteria were applied nationally and to subpopulations of the RPT_UNIT_6 or WETCLS_GRP
subpopulation groups, that is to:
• All Wetlands - Conterminous US (Table 8-1)
• RPT_UNIT_6 subpopulations: TDL, ICP, EMU, PLN-ARW, WVM (Table 8-3)
• WETCLS_GRP subpopulations: EH, EW, PRLH, PRLW (Table 8-4)
The metrics passing screening tests (range, repeatability, responsiveness criteria, and Step 1 of the
redundancy criteria) for a given subpopulation were retained for consideration in VMMI development.
8.6 Metric Screening Results
Candidate vegetation metrics that passed screening tests (Section 8.5) for the national scale, for five
subpopulations based on RPT_UNIT_6, or the subpopulations of WETCLS_GRP were retained for further
analysis. Passing metrics for each subpopulation were used in developing potential VMMIs for that
subpopulation. In the VMMI development process (described in Chapter 9:), four final VMMIs were
ultimately selected as the best performing, one for each WETCLS_GRP subpopulation: Estuarine
Herbaceous (EH), Estuarine Woody (EW), Inland Herbaceous (PRLH), and Inland Woody (PRW). These
Wetland Group VMMIs were used for population estimates of condition for the 2016 survey and for
change analysis between 2011 and 2016. Therefore, in this section we report metric screening results only
for the Wetland Group subpopulations (Table 8-6 through Table 8-9).
Table 8-6. Metrics (n = 40) that passed screening criteria for the Estuarine Herbaceous (EH) wetland
subpopulation. Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section 8.8 (Appendix E).
Estuarine Herbaceous
Range
S:N
Chi
P
Box
Metric Type
Wetland (EH) Metrics
Test
Ratio
Square
Value
plot
Score
TOTN_SPP
PASS
26.18
41.15
0.0000
2
All or Native Species
TOTN_FAM
PASS
24.3
34.87
0.0000
2
All or Native Species
H_ALL
PASS
47.24
29.66
0.0000
2
All or Native Species
XBCDIST_SPP
PASS
21.6
26.99
0.0000
2
All or Native Species
TOTN_NATSPP
PASS
29.6
30.05
0.0000
1
All or Native Species
PCTN_NATSPP
PASS
18.26
59.64
0.0000
1
All or Native Species
rfreclnatspp
PASS
27.6
64.49
0.0000
1
All or Native Species
H_NAT
PASS
18.98
25.23
0.0000
2
All or Native Species
XBCDIST_NATSPP
PASS
16.39
27.64
0.0000
2
All or Native Species
XC_NAT
PASS
57.23
18.79
0.0000
1
Floristic Quality
XC_ALL
PASS
60.34
40.06
0.0000
2
Floristic Quality
FQAI_COV_NAT
PASS
11.87
45.56
0.0000
2
Floristic Quality
FQAI_COV_ALL
PASS
14.06
61.41
0.0000
2
Tolerance
N_TOL
PASS
25.31
43.8
0.0000
2
Tolerance
126
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Estuarine Herbaceous
Wetland (EH) Metrics
Range
Test
S:N
Ratio
Chi
Square
P
Value
Box
plot
Score
Metric Type
PCTN_SEN
PASS
38.06
39.08
0.0000
2
Tolerance
PCTN_TOL
PASS
59.19
49.27
0.0000
2
Tolerance
XRCOV_SEN
PASS
115.37
46.9
0.0000
2
Tolerance
XRCOV_TOL
PASS
56.52
57.21
0.0000
2
Tolerance
XRCOV_HTOL
PASS-
21.98
56.07
0.0000
1
Tolerance
PCTN_OBL
PASS
14.82
41.74
0.0000
2
Hydrophytic Status
PCTN_OBL_FACW
PASS
37.3
35.26
0.0000
1
Hydrophytic Status
XRCOV_OBL
PASS
67.43
27.68
0.0000
1
Hydrophytic Status
WETIN D_COV_ALL
PASS
39.45
31.28
0.0000
2
Hydrophytic Status
WETIN D2_COV_ALL
PASS-
39.45
31.28
0.0000
2
Hydrophytic Status
N_FORB
PASS
22.15
55.93
0.0000
2
Growth Habit
XRCOV_FORB
PASS
79.94
41.43
0.0000
2
Growth Habit
PCTN_GRAMINOID_NAT
PASS
8.04
37.81
0.0000
2
Growth Habit
XRCOV_GRAMINOID_NAT
PASS
35.55
46.85
0.0000
2
Growth Habit
N_HERB
PASS
24.94
48.72
0.0000
2
Growth Habit
XRCOV_HERB_NAT
PASS
22.74
24.26
0.0000
2
Growth Habit
N_ANNUAL
PASS-
3.08
42.3
0.0000
1
Duration
N_PERENNIAL
PASS
20
35.05
0.0000
1
Duration
N_PERENNIAL_NAT
PASS
20.44
26.05
0.0000
1
Duration
PCTN_PERENNIAL_NAT
PASS
8.27
62.92
0.0000
2
Duration
N_DICOT
PASS
23.02
39.67
0.0000
2
Category
N_MONOCOT
PASS
13.63
27.1
0.0000
2
Category
PCTN_MONOCOTS_NAT
PASS
7.61
37.05
0.0000
2
Category
XRCOV_DICOT
PASS
44.16
28.45
0.0000
2
Category
XRCOV_MONOCOT
PASS
40.53
28.13
0.0000
2
Category
XRCOV_MONOCOTS_NAT
PASS
36.39
45.87
0.0000
2
Category
Table 8-7. Metrics (n = 21) that passed screening criteria for the Estuarine Woody (EW) wetland subpopulation.
Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section 8.8 (Appendix E).
Estuarine Woody Wetland
(EW) Metrics
Range
Test
S:N
Ratio
Chi
Square
p Value
Box plot
Score
Metric Type
XTOTABCOV
PASS
2.74
4.14
0.0419
2
All or Native Species
PCTN_NATSPP
PASS
9.15
7.44
0.0064
2
All or Native Species
RIMP_NATSPP
PASS-
122.29
8.97
0.0027
2
All or Native Species
FQAI_ALL
PASS
17.73
3.57
0.0587
1
Floristic Quality
PCTNJSEN
PASS
6.26
4.2
0.0405
2
Tolerance
PCTN_HTOL
PASS
22.33
4.47
0.0345
1
Tolerance
PCTN_GRAMINOID_NAT
PASS
10.68
8.38
0.0038
2
Graminoid
XABCOV_GRAMINOID
PASS
61.84
5.21
0.0225
2
Graminoid
XABCOV_GRAMINOID_NAT
PASS
115.16
7.34
0.0068
2
Graminoid
XRCOV_GRAMINOID
PASS
72.74
4.07
0.0436
2
Graminoid
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Estuarine Woody Wetland
Range
S:N
Chi
p Value
Box plot
Metric Type
(EW) Metrics
Test
Ratio
Square
Score
XRCOV_GRAMINOID_NAT
PASS
44.94
5.85
0.0156
2
Graminoid
XCOV_WD_FINE
PASS
40.88
4.69
0.0303
2
Woody
XRCOV_SHRUB_COMB
PASS
80.27
3.63
0.0568
1
Woody
XRCOV_SHRUB_COMB_NAT
PASS
80.76
4.04
0.0445
1
Woody
PCTN_DICOT
PASS
13.93
4.1
0.0429
2
Dicots
XRCOV_DICOT
PASS
74.9
4.31
0.0378
1
Dicots
XRCOV_DICOTS_NAT
PASS
70.83
3.58
0.0584
1
Dicots
PCTN_MONOCOT
PASS
8.86
5.91
0.015
2
Monocots
PCTN_MONOCOTS_NAT
PASS
7.94
9.77
0.0018
3
Monocots
XABCOV_MONOCOT
PASS
67.64
7.62
0.0058
3
Monocots
XRCOV_MONOCOTS_NAT
PASS
49.56
8.34
0.0039
3
Monocots
Table 8-8. Metrics (n = 42) that passed screening criteria for the Inland Herbaceous (PRLH) wetland subpopulation.
Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section 8.8 (Appendix E).
Inland Herbaceous Wetland
Range
S:N
Chi
p Value
Box
Metric Type
(PRLH) Metrics
Test
Ratio
Square
plot
Score
PCTN_NATSPP
PASS
8.14
79.11
0.0000
3
Native Species
rfreclnatspp
PASS
10.61
82.2
0.0000
3
Native Species
XRCOV_NATSPP
PASS
8.37
86.26
0.0000
3
Native Species
RIMP_NATSPP
PASS
11.84
92.83
0.0000
3
Native Species
XC_NAT
PASS
23.06
62.01
0.0000
2
Floristic Quality
XC_ALL
PASS
37.2
94.31
0.0000
3
Floristic Quality
XC_COV_ALL
PASS
28.62
45.66
0.0000
2
Floristic Quality
FQAI_NAT
PASS
30.81
30.74
0.0000
2
Floristic Quality
FQAI_ALL
PASS
38.13
40.38
0.0000
2
Floristic Quality
FQAI_COV_ALL
PASS
14.7
69.77
0.0000
2
Floristic Quality
N_SEN
PASS
25.5
38.99
0.0000
1
Sensitive
PCTN_SEN
PASS
18.17
53.79
0.0000
2
Sensitive
PCTN_ISEN
PASS
9.09
24.95
0.0000
2
Sensitive
XRCOV_SEN
PASS
24.21
37.16
0.0000
1
Sensitive
XRCOV_ISEN
PASS
7.32
24.08
0.0000
1
Sensitive
N_TOL
PASS
8.85
35.94
0.0000
1
Tolerant
N_HTOL
PASS
6.51
58.06
0.0000
2
Tolerant
PCTN_TOL
PASS
17.36
66.16
0.0000
2
Tolerant
PCTN_HTOL
PASS
18.1
71.48
0.0000
3
Tolerant
XRCOV_TOL
PASS
8.68
57.34
0.0000
2
Tolerant
XRCOV_HTOL
PASS
12.83
66.44
0.0000
2
Tolerant
PCTN_FAC_FACU
PASS
11.77
47.19
0.0000
2
Hydrophytic Status
PCTN_OBL_FACW
PASS
20.25
47.64
0.0000
2
Hydrophytic Status
PCTN_OBL_FACW_FAC
PASS
9.95
33.08
0.0000
2
Hydrophytic Status
XRCOV_OBL
PASS
23.08
51.97
0.0000
2
Hydrophytic Status
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Inland Herbaceous Wetland
(PRLH) Metrics
Range
Test
S:N
Ratio
Chi
Square
p Value
Box
plot
Score
Metric Type
XRCOV_FAC_FACU
PASS
9.23
42.47
0.0000
2
Hydrophytic Status
XRCOV_OBL_FACW
PASS
25.31
40.06
0.0000
2
Hydrophytic Status
WETIN D2_COV_ALL
PASS-
30.75
51.2
0.0000
2
Hydrophytic Status
WETIN D2_COV_N AT
PASS-
15.29
30.7
0.0000
1
Hydrophytic Status
PCTN_FORB_NAT
PASS
6.7
28.73
0.0000
2
Herbaceous
XRCOV_GRAMINOID_NAT
PASS
17.3
12.02
0.0005
0
Herbaceous
PCTN_HERB_NAT
PASS
6.63
30.75
0.0000
1
Herbaceous
XRCOV_HERB_NAT
PASS
8.98
41.62
0.0000
2
Herbaceous
PCTN_SHRUB_COMB
PASS
19.96
17.36
0.0000
1
Shrub
PCTN_SHRUB_COMB_NAT
PASS
14.52
16.81
0.0000
1
Shrub
XRCOV_SHRUB_COMB
PASS
24.64
10.57
0.0011
0
Shrub
XRCOV_SHRUB_COMB_NAT
PASS
24.57
9.87
0.0017
0
Shrub
PCTN_ANNUAL
PASS
7.8
14.51
0.0001
0
Category
PCTN_PERENNIAL
PASS
11.06
24.6
0.0000
2
Category
PCTN_PERENNIAL_NAT
PASS
13.41
60.92
0.0000
2
Category
XRCOV_PERENNIAL_NAT
PASS
10.8
62.87
0.0000
2
Category
XRCOV_MONOCOTS_NAT
PASS
9.04
18.53
0.0000
1
Category
Table 8-9. Metrics (n = 47) that passed screening criteria for the Inland Woody (PRLW) wetland subpopulation.
Kruskal-Wallis statistics: Chi square and p-value. Metrics defined in Section 8.8 (Appendix E).
Inland Woody Wetland
(PRLW) Metrics
Range
Test
S:N
Ratio
Chi
Square
p Value
Box
plot
Score
Metric Type
PCTN_NATSPP
PASS
7.11
51.39
0.0000
2
Native Species
rfreclnatspp
PASS
12.63
56.12
0.0000
2
Native Species
XRCOV_NATSPP
PASS-
18.29
65.53
0.0000
2
Native Species
RIMP_NATSPP
PASS
20.14
64.77
0.0000
2
Native Species
XC_NAT
PASS
49.34
27.37
0.0000
1
Floristic Quality
XC_ALL
PASS
62.91
47.61
0.0000
2
Floristic Quality
FQAI_COV_ALL
PASS
49.62
30.81
0.0000
0
Floristic Quality
PCTN_SEN
PASS
34.53
37.56
0.0000
1
Tolerance
PCTN_TOL
PASS
39.68
32.01
0.0000
1
Tolerance
PCTN_HTOL
PASS
28.94
35.93
0.0000
1
Tolerance
XRCOV_HTOL
PASS
25.05
37.11
0.0000
1
Tolerance
PCTN_FAC_FACU
PASS
13.71
12.09
0.0005
0
Hydrophytic Status
PCTN_OBL_FACW
PASS
16.23
18.3
0.0000
0
Hydrophytic Status
XRCOV_UPL
PASS
35.16
11.68
0.0006
0
Hydrophytic Status
XRCOV_FAC_FACU
PASS
20.36
11.47
0.0007
0
Hydrophytic Status
XRCOV_OBL_FACW
PASS
18.13
15.56
0.0001
0
Hydrophytic Status
WETI N D2_COV_ALL
PASS
24.24
14.54
0.0001
0
Hydrophytic Status
PCTN_HERB
PASS
24.74
6.38
0.0115
0
Vine
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Inland Woody Wetland
Range
S:N
Chi
p Value
Box
Metric Type
(PRLW) Metrics
Test
Ratio
Square
plot
Score
PCTN_VINE_ALL
PASS
16.29
7.5
0.0062
0
Vine
XRCOV_VINE_ALL
PASS
28.36
9.92
0.0016
0
Vine
XRCOV_VI NE_ALL_NAT
PASS
30.92
8.26
0.0041
0
Vine
PCTN_SHRUB_COMB
PASS
32.05
18.1
0.0000
0
Shrub
PCTN_SHRUB_COMB_NAT
PASS
11.11
10.44
0.0012
0
Shrub
XRCOV_SHRUB_COMB_NAT
PASS
41.98
26.03
0.0000
0
Shrub
PCTN_TREE_COMB
PASS
25.43
8.25
0.0041
0
Tree
XRCOV_TREE_COMB_NAT
PASS
20.59
5.72
0.0168
0
Tree
XRCOV_GYMNOSPERM
PASS
24.45
20.89
0.0000
1
Tree
IMP_TREE_GROUND
PASS
2.12
7.15
0.0075
0
Tree
IMP_TREE_UPPER
PASS
6.23
7.01
0.0081
0
Tree
TOTN_TREES
PASS
10.42
11.63
0.0007
0
Tree
TOTN_MID
PASS
11.59
7.88
0.005
0
Tree
TOTN_SMALL
PASS
9.24
10.81
0.001
0
Tree
TOTN_SNAGS
PASS
11.96
20.25
0.0000
0
Tree
XN_SNAGS
PASS
11.94
20.39
0.0000
0
Tree
PCTN_PERENNIAL
PASS
7.99
38.6
0.0000
2
Duration
PCTN_PERENNIAL_NAT
PASS
9.29
56.34
0.0000
2
Duration
XRCOV_ANNUAL
PASS
29.94
11.47
0.0007
0
Duration
XRCOV_ANNUAL_NAT
PASS
31.48
11.77
0.0006
0
Duration
XRCOV_PERENNIAL_NAT
PASS
24.97
56.17
0.0000
1
Duration
XRCOV_MONOCOTS_NAT
PASS
14.99
8.07
0.0045
0
Duration
PCTN_FERN
PASS
15.36
10.11
0.0015
0
Non-seed Plants
PCTN_FERNS_NAT
PASS
14.95
10.86
0.001
0
Non-seed Plants
XRCOV_FERN
PASS
14.56
7.86
0.005
0
Non-seed Plants
freclbryophytes
PASS
2.42
29.2
0.0000
1
Non-seed Plants
IMP_BRYOPHYTES
PASS
5.88
26.81
0.0000
0
Non-seed Plants
XCOV_LI CHENS
PASS
4.26
33.36
0.0000
1
Non-seed Plants
IMP_LICHENS
PASS
5.53
13.91
0.0002
0
Non-seed Plants
8.7 Literature Cited
Barbour MT, Gerritsen J, Griffth GE, Frydenborg R, McCarron E, White JS, Bastian ML (1996) A framework
for biological criteria for Florida streams using benthic macroinvertebrates. Journal of North American
Benthological Society 15: 185-211
Bates D, Maechler M, Bolker B and Walker S (2014) Ime4: Linear mixed-effects models using Eigen and S4
R package version 1.1-7 (http://CRAN.R-project.org/packageHme4).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
130
-------
Herlihy AT, Kentula ME, Magee TK, Lomnicky GA, Nahlik AM, Serenbetz G (2019) Strivingfor consistency
in the National Wetland Condition Assessment: developing a reference condition approach for assessing
wetlands at a continental scale. Environmental Monitoring and Assessment 191 (1):327.
doi: 10.1007/s 10661-019-7325-3
Jongman RHG, Ter Braak CJF, van Tongeren OFR (eds) (1995) Data Analysis in Community and Landscape
Ecology. Cambridge University Press, Cambridge, United Kingdom
Kaufmann PR, Levine P, Robison EG, Seeliger C, Peck DV (1999) Quantifying Physical Habitat in Wadable
Streams. EPA/620/R_99/003. US Environmental Protection Agency, Washington, DC
Klemm DJ, Blocksom KA, Thoeney WT, FulkFA, HerlihyAT, Kaufmann PR, CorimerSM (2002) Methods
development and use of macroinvertebrates as indicators of ecological conditions for streams in the Mid-
Atlantic Highlands Region. Environmetal Monitioring and Assessment 78: 169-212
Kuebbing SE, Classen AT, Sanders NJ, Simberloff D (2015) Above- and below-ground effects of plant
diversity depend on species origin: an experimental test with multiple invaders. New Phytologist 208
(3):727-735. doi:10.1111/nph.13488
Magee TK, Ringold PL, Bollman MA (2008) Alien species importance in native vegetation along wadeable
streams, John Day River basin, Oregon, USA. Plant Ecology 195:287-307. doi:10.1007/sll258-007-9330-9
Magee TK, Ringold PL, Bollman MA, Ernst TL (2010) Index of Alien Impact (IAI): A method for evaluating
alien plant species in native ecosystems. Environmental Management 45:759-778. doi:10.1007/s00267-
010-9426-1
Magee TK, Blocksom KA, Fennessy MS (2019a) A national-scale vegetation multimetric index (VMMI) as
an indicator of wetland condition across the conterminous United States. Environmental Monitoring and
Assessment 191 (SI): 322, doi: 10.1007/sl0661-019-7324-4.
https://link.springer.com/article/10.1007/sl0661-019-7324-4
Magee TK, Blocksom KA, HerlihyAT, &. Nahlik AM (2019b) Characterizing nonnative plants in wetlands
across the conterminous United States. Environmental Monitoring and Assessment 191 (SI): 344, doi:
10.1007/s 10661-019-7317-3. https://link.springer.com/article/10.1007/sl0661-019-7317-3
McCune, B and Mefford, MJ (2018) PC-ORD. Multivariate Analysis of Ecological Data.Version 7.08
McCune B, Grace JB (2002) Analysis of Ecological Communities. Gleneden Beach, OR: MjM Software
Design
Pont D, Hughes RM, Whittier TR, Schumtz S (2009) A predictive index of biotic integrity model for aquatic-
vertebrate assemblages of western US streams. Transactions of the American Fisheries Society 138: 292-
305
Pysek P, Hulme P, Simberloff D, Bacher S, Blackburn T, Carlton J, Foxcroft L, Genovesi P, Jeschke J, Kuhn I,
Liebhold A, Mandrak N, Meyerson L, Pauchard A, Pergl J, Roy H, Richardson D (2020) Scientists' warning
on invasive alien species. Biological Reviews. doi:10.1111/brv.12627
131
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
R Core Team (2019) R: A language and environment for statistical computing. Version 3.6.1. R Foundation
for Statistical Computing, Vienna, Austria. (http://www.R-project.org/)
Ricciardi A, lacarella JC, Aldridge DC, Blackburn TM, Carlton JT, Catford JA, Dick JTA, Hulme PE, Jeschke
JM, Liebhold AM, Lockwood JL, Maclsaac HJ, Meyerson LA, Pysek P, Richardson DM, Ruiz GM, Simberloff
D, Vila M, Wardle DA (2020) Four priority areas to advance invasion science in the face of rapid
environmental change. Environmental Reviews:l-23. doi:10.1139/er-2020-0088
Ruaro R, Gubiani EA, Thomaz SM, Mormul RP (2020) Nonnative invasive species are overlooked in
biological integrity assessments. Biological Invasions. doi:10.1007/sl0530-020-02357-8
Simberloff D (2011) How common are invasion-induced ecosystem impacts? Biological Invasions 13
(5):1255-1268. doi:10.1007/sl0530-011-9956-3
Stoddard JL, Herlihy AT, Peck DV, Hughes RM, Whittier TR, Tarquinio E (2008) A process for creating
multimetric indices for large-scale aquatic surveys. Journal of North American Benthological Society 27:
878-891
USEPA (2006) Wadeable Streams Assessment: A Collaborative Survey of the Nation's Streams. EPA 841-B-
06-002. US Environmental Protection Agency, Washington, DC
USEPA (2016a) National Wetland Condition Assessment: 2011 Technical Report. EPA-843-R-15-006. .US
Environmental Protection Agency, Washington, DC. https://www.epa.gov/national-aquatic-resource-
surveys/national-wetland-condition-assessment-2011-results
Van Sickle J (2010) Correlated metrics yield multimetric indices with inferior performance. Transactions of
the American Fisheries Society 139: 1802-1817
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
132
-------
8.8 Appendix E: NWCA Candidate Vegetation Metrics4
READ THIS: Key Information for Reading and Using This Appendix
• Important: This Appendix is a descriptive overview of Candidate Vegetation Metrics. Exact methods/formulas
for calculations, specific field data, and trait information used for each metric were defined in the Vegetation
Metric R Code.
• Unless otherwise indicated, vegetation metrics are summarized to site level. Metrics are calculated based on
data from five 100-m2 plots in the Assessment Area (AA) for the site (or if fewer than 5 plots were sampled,
the total number plots sampled). In the metric descriptions or formulas provided in this appendix, the phrase
'five 100-m2 plots' can be assumed to mean the 5 plots in the AA or the total number of plots sampled if less
than 5. Rarely were fewer than 5 vegetation plots sampled at the AA.
• The term 'Species' as typically used in this appendix refers to taxonomic species or lowest identifiable
taxonomic unit (e.g., variety, genus, family, growth-habit).
• BLACK BANNER with column headings is repeated at the top of each page.
• GRAY BANNER, heading each major group of metrics, lists the NWCA Field Data Form from which the
validated field data that is used in metrics originated.
• COLORED BANNERS, under each major metric group, provide section and subsection headings for sets of
metrics that describe related ecological components.
• METRIC NAME column the metric name used in the NWCA vegetation metrics data set.
• DESCRIPTION column gives narrative description of each metric.
• CALCULATION/TRAIT INFORMATION column provides:
o In white metric rows:
* A general formula for calculation of the metric, if not evident in the DESCRIPTION column, is provided.
PARAMETER NAMES representing raw data included in calculations are highlighted in GRAY-BLUE and
re defined in Section 5.12, Appendix C.
¦ Some calculated metrics listed in the METRIC NAME column are, in turn, used as components of other
calculated metrics.
¦ Some calculated metrics use species trait information to aggregate species level data. Where traits are
used, trait names are indicated in the calculation column using GREEN font.
o In colored banner rows defining metric sets - General categories of species trait information used in
calculating a particular series of metrics are listed, if applicable. Codes for specific traits are indicated in
GREEN font. For metrics that use species traits, trait designations are applied as follows:
* Growth-habit, Duration, and Taxonomic Category are applied by species (see Section 5.6)
¦ Wetland Indicator Status is applied to taxon-wetland region pairs (see Section 5.7)
¦ Native status designations for taxon-site pairs are based on state-level status (see Section 5.8)
¦ Coefficients of Conservatism (CCs, aka C-values) are applied to taxon-site pairs based on state or
regional specific C-values for each species (see Section 5.9)
• METRIC TYPE column indicates whether the candidate metric is to reflect ecological condition or stress.
• METRICS INCLUDED IN NWCA VEGETATION INDICES are indicated in the METRIC TYPE column in bold color-
coded font: the four 2016 Wetland Type Vegetation Multimetric Indices (VMMIs) in light blue (EH), dark blue
(EW), purple (PRLH), forest green (PRLW), respectively; the Nonnative Plant Indicator (NNPI) in red; and the
previously used 2011 National (VMMI) in "ose.
4 Most metrics developed for analysis of the 2011 NWCA vegetation data (USEPA 2016a) were considered here. A few (n = 11)
metrics were dropped because the 2016 field protocols were simplified and requisite data for those specific metrics were
unavailable for 2016 data. Also, several new metrics that described additional characteristics of hydrophytic vegetation (n =16),
vines (n = 12), and summaries of tree counts by three major size (dbh) ranges (n = 6) were added.
133
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
SECTIONS 1- 5
Metrics based on field data: FORM V-2 - NWCA 2016 VASCULAR
SPECIES PRESENCE AND COVER
SECTION 1
TAX A COMPOSITION (RICHNESS,
FREQUENCY, COVER, DIVERSITY)
Section 1.1
All Species/Taxonomic Groups
TOTN_SPP
Richness - Total number of unique
Count unique species across all
r
species across all 100-m2 plots
plots
XN_SPP
Mean number of species across all
100-m2 plots
C
MEDN_SPP
Median number of species across all
100-m2 plots
C
SDN_SPP
Standard deviation in number of
species across all 100-m2 plots
C
TOTN_GEN
Total number of unique genera
Count unique genera across all
r
across all 100-m2 plots
plots
XN_GEN
Mean number of unique genera
across all 100-m2 plots
C
MEDN_GEN
Median number of genera across all
100-m2 plots
C
SDN_GEN
Standard deviation in number of
genera across 100-m2 plots
C
TOTN FAM
Total number of families across
Count unique families observed
c
100-m2 plots
across all plots
L
XN_FAM
Mean number of families across
100-m2 plots
C
MEDN_FAM
Median number of families across
100-m2 plots
C
SDN_FAM
Standard deviation in number of
families across 100-m2 plots
C
XTOTABCOV
Mean total absolute cover summed
S COVER of II individual taxa
(summary data
across all species across 100-m2
across 5 plots/5 plots
used in calculation
plots
of other metrics)
H_ALL
Shannon-Wiener Diversity Index -
All species
s = number of species observed, / =
species i, p = proportion of
individuals (relative cover)
belonging to species /
3
H* =
£
C
J_ALL
Evenness (Pielou) - All species
H"
' ~ his
S = number of species observed
C
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
134
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
D_ALL
Simpson Diversity Index - All species
3
s = number of species observed, / =
D = 1 "Xrf
c
species i, p = proportion of
individuals (relative cover)
belonging to species /
XBCDIST_SPP
Within Assessment Area
dissimilarity based on species
composition = Mean of between-
plot Bray-Cutis (BC) Distance
Calculate between-plot Bray Curtis
Distance for all plot pairs based on
species and plot level cover
values. Calculate mean of these
(Dissimilarity) based on all species.
values to get mean within AA
distance:
2 V* MlN(atpahJ)
BCih = 1 J
Zj=iaU+ ^j=iahj
C
SECTIONS 1.2 -1.3
NATIVE STATUS
Trait Information = Native Status
(see Table 5-5)
Section 1.2
Native ( J/5 ) Species/Taxonomic
Groups
TOTN_NATSPP
Native Richness: Total number of
Count unique native (NAT) species
unique native species across all 100-
across all plots
C
m2 plots
XN_NATSPP
Mean number of native species
across 100-m2 plots
C
MEDN_NATSPP
Median number of native species
across 100-m2 plots
C
SDN_NATSPP
Standard deviation in number of
native species across 100-m2 plots
C
PCTN_NATSPP
Percent richness of native species
(TOTN_NATSPP/TOTN_SPP) x 100
c,
observed across 100-m2 plots
in EH-VVMI,
EW-VMMI
RFREQJMATSPP
Relative frequency of occurrence
2 Frequencies of all (NAT
for native species as a percent of
species/2 Frequencies of all
C, in
PRLW-VMMI
total frequency (sum of all species)
species) x 100; Frequency for
individual species = % of 100-m2
plots in which it occurs.
XABCOV
Mean total absolute cover of native
2 COVER of all individual native
c
NATSPP
species across 100-m2 plots
(NAT) taxa across 5 plots/5 plots
XRCOV_NATSPP
Mean relative cover of native
(XABCOV_NATSPP/XTOTABCOV) x
C, in
species across 100-m2 plots as a
percentage of total cover
100
PRLH-VMMI,
PRLW-VMMI
RIMP_NATSPP
Mean relative importance of all
(RFREQJMATSPP +
c,
¦, ¦
native species
XRCOV_NATSPP)/2
in EW-VMMI,
2011 National
VMMI
H_NAT
Shannon-Wiener Diversity Index -
Native species only
See H_ALL
C
135
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
J_NAT
Evenness (Pielou) - Native species
only
See J_ALL
C
D_NAT
Simpson Diversity Index - Native
species only
See D_NAT
C
XBCDIST_
Within AA dissimilarity based on
See XBCDIST_SPP
NATSPP
native species only composition =
Mean of between plot Bray-Cutis
Distance (Dissimilarity) based on
native species only
C
Section 1.3
Introduced ( MTR), Adventive
( ), M.IEN (INTR + ADV),
Cryptogenic (CRY )
Trait Information = Native Status
(see Table 5-5)
TOTNJNTRSPP
Introduced Richness: Total number
Count unique introduced (INTR)
of unique introduced species across
species across all plots
S
all 100-m2 plots
XN_INTRSPP
Mean number of introduced species
across 100-m2 plots
S
MEDNJNTRSPP
Median number of introduced
species across 100-m2 plots
S
SDNJNTRSPP
Standard deviation in number of
introduced species across 100-m2
plots
S
PCTN_INTRSPP
Percent richness introduced species
observed across 100-m2 plots
(TOTN_INTRSPP/TOTN_SPP) x 100
S
rfreqjntrspp
Relative frequency of occurrence
for introduced species as a percent
(2 Frequencies of all introduced
(INTR) species/2 Frequencies of all
of total frequency (sum of all
species) x 100; Frequency for
S
species)
individual species = % of 100-m2
plots in which it occurs.
XABCOV_
Mean total absolute cover of all
S COVER of all individual INTR
INTRSPP
introduced species across 100-m2
plots
taxa across 5 plots/5 plots
S
XRCOV_INTRSPP
Mean relative cover of all INTR
(XABCOVJNTRSPP/XTOTABCOV) x
species across 100-m2 plots as a
100
S
percentage of total cover
RIMPJNTRSPP
Mean relative importance of all
(RFRECLINTRSPP+
c
introduced species
XRCOV_INTRSPP)/2
J
TOTN_ADVSPP
Adventive Richness: Total number
Count unique adventive (ADV)
of adventive species across 100-m2
species across all plots
s
plots
XN_ADVSPP
Mean number of adventive species
across 100-m2 plots
s
MEDN_ADVSPP
Median number of adventive
species across 100-m2 plots
s
SDN_ADVSPP
Standard deviation in number of
adventive species across 100-m2
plots
s
136
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric METRIC TYpE
I \uvv/; . .
/.«p«p^TrN.1T.p»mp /¦ ^ (C = condition,
SPECIES TRAIT TYPE (indicated in
METRIC NAME METRIC DESCRIPTION Banner if applicable) -stress)
PCTN_ADVSPP
Percent richness adventive species
observed across all 100-m2 plots
(TOTN_ADVSPP/TOTN_SPP) x 100
S
rfrecladvspp
Relative frequency of adventive
species occurrence across 100-m2
(2 Frequencies of all adventive
(ADV) species/2 Frequencies of all
plots
species) x 100; Frequency for
individual species = % of 100-m2
plots in which it occurs.
S
XABCOV_
Mean total absolute cover of all
S COVER of all individual ADV taxa
ADVSPP
ADV species across 100-m2 plots
across 5 plots/5 plots
XRCOV_ADVSPP
Mean relative cover of all ADV
(XABCOV_ADVSPP/XTOTABCOV) x
species or lowest taxonomic unit
100
across 100-m2 plots as a percentage
J
of total cover
RIMP_ADVSPP
Mean relative importance of all
(RFRECLADVSPP +
C
adventive species
XRCOV_ADVSPP)/2
3
TOTN_ALIENSPP
Alien Richness: Total number of
unique alien (INTR + ADV) species
across 100-m2 plots
TOTN_ADVSPP + TOTNJNTRSPP
s
XN_ALIENSPP
Mean number of alien (INTR + ADV)
species across 100-m2 plots
s
MEDN_ALIENSPP
Median number of alien (INTR +
ADV) species across 100-m2 plots
s
SDN_ALIENSPP
Standard deviation in number of
alien (INTR + ADV) species
s
PCTN_ALIENSPP
Percent richness alien species
(TOTN_ALIENSPP/TOTN_SPP) x
c
across 100-m2 plots
100
RFRECL
Relative frequency of alien (INTR +
(2 Frequencies of all ALIEN
AUENSPP
ADV) species occurrence across
species/2 Frequencies of all
100-m2 plots
species) x 100; Frequency for
individual species = % of 100-m2
plots in which it occurs.
s
XABCOV_
Mean total absolute cover of ALIEN
S COVER of all individual ALIEN
AUENSPP
(INTR + ADV) species across 100-m2
plots
taxa across 5 plots/5 plots
s
XRCOV_
Mean relative cover of all ALIEN
(XABCOV_ALIENSPP/XTOTABCOV)
AUENSPP
(INTR + ADV) species across 100-m2
plots as a percentage of total cover
x 100
s
RIMP_ALIENSPP
Mean relative importance of all
(RFRECLALIENSPP +
c
ALIEN (INTR + ADV) species
XRCOV_ALIENSPP)/2
J
H_ALIEN
Shannon-Wiener Diversity Index
See H_ALL
s
J_ALIEN
Evenness (Pielou)
See J_ALL
s
D_ALIEN
Simpson Diversity Index
See D_NAT
s
TOTN_CRYPSPP Cryptogenic Richness: Total number Count unique cryptogenic (CRYP)
of unique cryptogenic species species across all plots S
across 100-m2 plots
XN_CRYPSPP Mean number of cryptogenic
species across 100-m2 plots
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
MEDN_CRYPSPP
Median number of cryptogenic
species across 100-m2 plots
S
SDN_CRYPSPP
Standard deviation in number of
cryptogenic species across 100-m2
plots
S
PCTN_CRYPSPP
Percent richness cryptogenic
species across 100-m2 plots
(TOTN_CRYPSPP/TOTN_SPP) x 100
S
RFREQ_CRYPSPP Relative frequency of cryptogenic (£ Frequencies of all cryptogenic
species occurrence across 100-m2 (CRYP) species/2 Frequencies of
plots all species) x 100; Frequency for S
individual species = % of 100-m2
plots in which it occurs.
XABCOV_
Mean total absolute cover of all
S COVER of all CRYP taxa across 5
C
CRYPSPP
CRYP species across 100-m2 plots
plots/5 plots
J
XRCOV_CRYPSPP
Mean relative cover of all CRYP
(XA BCO V_C RYPS P P/XTOTA BCO V)
species across 100-m2 plots as a
x 100
s
percentage of total cover
RIMP_CRYPSPP
Mean relative importance of all
(RFRECLCRYPSPP+
c
CRYP species
XRCOV_CRYPSPP)/2
J
TOTN_AC
AC Richness: Total number of
TOTN_CRYPSPP + TOTN_ALIENSPP
S, Used in
NNPI
unique alien and cryptogenic
species across 100-m2 plots
XN_AC
Mean number of AC (ALIEN + CRYP)
species across 100-m2 plots
S
MEDN_AC
Median number of AC (ALIEN +
CRYP) species across 100-m2 plots
S
SDN_AC
Standard deviation number of AC
(ALIEN + CRYP) species across 100-
m2 plots
s
PCTN_AC
Percent Richness AC species (ALIEN
(TOTN_CRYPSPP + TOTN-
c
+ CRYP) across 100-m2 plots
ALIENSPP/TOTN_SPP) x 100
J
RFRECLAC
Relative frequency of alien and
(2 Frequencies of all ALIEN + CRYP
cryptogenic species occurrence in
species/2 Frequencies of all
S, Used in
NNPI
flora based on five 100-m2 plots
species) x 100; Frequency for
individual species = % of 100-m2
plots in which it occurs.
XABCOV_AC
Mean total absolute cover of all AC
S COVER of all ALIEN + CRYP taxa
(ALIEN + CRYP) species across 100-
across 5 plots/5 plots
S
m2 plots
XRCOV_AC
Mean relative cover of all AC (ALIEN
(XA BCOV_AC/XTOTABCOV) x 100
S, Used in
NNPI
+ CRYP) species across 100-m2 plots
as a percentage of total cover
RIMP_AC
Mean relative importance of all AC
(ALIEN + CRYP) species
(RFRECLAC + XRCOV_AC)/2
S
H_AC
Shannon-Weiner Diversity Index
See H_ALL
S
J_AC
Evenness (Pielou)
See J_ALL
s
D_AC
Simpson Diversity Index
See D_NAT
s
¦
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
Section 2
FLORISTIC QUALITY
Trait Information =
Coefficients of Conservatism
(see Section 5.9); Native
Status (see Table 5-5)
Equation 1
General formula for Mean C
c = ^cc)/N_,
CC,y-coefficient of conservatism for
each unique species / at site j, N =
number of species at site j
Equation 2
General formula for FQAI
CC,y-coefficient of conservatism for
each unique species / at site j, N =
FOAI =
ica/jN,
number of species at site j
Equation 3
For weighted Mean C or FQAI
Replace CC,y with wCC,y where p,y =
relative frequency or relative cover
wCCij = PjjCC
J V
XC_NAT
Mean Coefficient of Conservatism
with native species only
Equation 1
C
XC_ALL
Mean Coefficient of Conservatism
with all species
Equation 1
C, in PRLW-
VMMI
xc_freclnat
Relative frequency-weighted Mean
Coefficient of Conservatism with
native species only
Equation 1, Equation 3
C
xc_freclaii
Relative frequency-weighted Mean
Coefficient of Conservatism with all
species only
Equation 1, Equation 3
c
XC_COV_NAT
Relative cover-weighted Mean
Coefficient of Conservatism with
native species only
Equation 1, Equation 3
c
XC_COV_AII
Relative cover-weighted Mean
Coefficient of Conservatism with all
species
Equation 1, Equation 3
c
FQAI_NAT
Floristic Quality Index with native
species only
Equation 2
c
FQAI.ALL
Floristic Quality Index with all
Equation 2
c, in PRLH-
species
VMMI, 2011
National
VMMI
FQAI_FRECLNAT
Proportional frequency-weighted
Floristic Quality Assessment Index
with native species only
Equation 2, Equation 3
C
FQAI_FRECLALL
Proportional frequency-weighted
Floristic Quality Assessment Index
with all species only
Equation 2, Equation 3
C
FQAI_COV_NAT
Proportional cover-weighted
Floristic Quality Assessment Index
with native species only
Equation 2, Equation 3
c
139
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
FQAI_COV_ALL
Proportional cover-weighted
Floristic Quality Assessment Index
with all species
Equation 2, Equation 3
C
Section 3
STRESS
TOLERANCE/SENSITIVITY
Trait Information =
Coefficients of Conservatism
(Section 5.9)
N_HSEN
Number (Richness) Highly Sensitive
Count unique species that meet
r
Species; C-value >= 9
criterion across 100-m2 plots
N_SEN
Number (Richness) Sensitive
Count unique species that meet
r
Species; C -value >= 7
criterion across 100-m2 plots
NJSEN
Number (Richness) Intermediate
Count unique species that meet
r
Sensitivity Species; C-value = 5 to 6
criterion across 100-m2 plots
N_TOL
Number (Richness) Tolerant
Count unique species that meet
C, in
Species; C -value <= 4
criterion across 100-m2 plots
PRLH-VMMI,
2011 National
VMMI
N_HTOL
Number (Richness) Highly Tolerant
Count unique species that meet
r
Species; C-value <= 2
criterion across 100-m2 plots
PCTN_HSEN
Percent Richness Highly Sensitive
Species; C-value >= 9
(N_HSEN/TOTN_SPP) x 100
C
PCTN_SEN
Percent Richness Sensitive Species;
C-value >= 7
(N_SEN/TOTN_SPP) x100
C
PCTN_ISEN
Percent Richness Intermediate
Sensitivity Species; C-value = 5 to 6
(N_ISEN/TOTN_SPP) x 100
C,
in EW-VMMI
PCTN_TOL
Percent Richness Tolerant Species;
C-value <= 4
(N_TO L/TOTN_S P P) x 100
C
PCTN_HTOL
Percent Richness Highly Tolerant
Species; C-value <= 2
(N_HTOL/TOTN_SPP) x 100
c
XABCOV_HSEN
Absolute Mean Cover Highly
S COVER of species with C-value
r
Sensitive Species; C-value >= 9
>= 9 across 5 plots/5 plots
L,
XABCOV SEN
Absolute Mean Cover Sensitive
S COVER of species with C-value
c
Species; C-value >= 7
>= 7 across 5 plots/5 plots
L
XABCOV ISEN
Absolute Mean Cover Intermediate
S COVER of species with C-value =
c
Sensitivity Species; C-value= 5 to 6
5 or 6 across 5 plots/5 plots
L
XABCOV_TOL
Absolute Mean Cover Tolerant
S COVER of species with C-value
r
Species; C-value <= 4
<= 4 across 5 plots/5 plots
L,
XABCOV_HTOL
Absolute Mean Cover Highly
S COVER of species with C-value
r
Tolerant Species; C-value <= 2
<= 2 across 5 plots/5 plots
L,
XRCOV_HSEN
Relative Mean Cover Highly
(XABCOV_HSEN/XTOTABCOV) x
r
Sensitive Species; C >= 9
100
XRCOV_SEN
Relative Mean Cover Sensitive
Species; C-value >= 7
(XA BCOV_S E N/XTOTABCOV) x 100
c,
in EH-VMMI
XRCOV_ISEN
Relative Mean Cover Intermediate
(XABCOVJSEN/XTOTABCOV) x
r
Sensitivity Species; C-value = 5 to 6
100
XRCOV_TOL
Relative Mean Cover Tolerant
Species; C-value <= 4
(XABCOV_TOL/XTOTABCOV) x 100
c
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
XRCOV_HTOL
Relative Mean Cover Highly
(XABCOV_HTOL/XTOTABCOV) x
c,
Tolerant Species; C-value <= 2
100
in EH-VMMI
SECTION 4
HYDROPHYTIC
CHARACTERISTICS OF
VEGETATION
Trait Information = Wetland
Indicator Status (WIS): Obligate
( 'Bl), Facultative Wetland
( OiC\l\ ), Facultative (FAC),
Facultative Upland ( ACL ), Upland
( PI) (Table 5-3); Native Status
(Table 5-5)
N_OBL
Richness (number) of Obligate
Count unique OBL species across
r
species
100-m2 plots
N_FACW
Richness (number) of Facultative
Count unique FACW species
r
Wetland species
across 100-m2 plots
N_FAC
Richness (number) of Facultative
Count unique FACU species across
r
species
100-m2 plots
N_FACU
Richness (number) of Facultative
Count unique FAC species across
r
Upland species
100-m2 plots
N_UPL
Richness (number) of UPL species =
Count unique UPL species across
r
UPL
100-m2 plots
N_OBL_FACW
Richness (number) of Obligate +
Count unique OBL+ FACW species
r
Facultative Wetland species
across 100-m2 plots
N_OBL_FACW_FAC
Richness (number) of Obligate +
Count unique OBL + FACW + FAC
r
Facultative Wetland species
species across 100-m2 plots
N_FAC_FACU
Richness (number) of Facultative +
Count unique FAC + FACU species
r
Facultative Upland species
across 100-m2 plots
PCTN_OBL
Percent richness of Obligate species
(N_OBL/TOTN_SPP) x 100
c
PCTN_FACW
Percent richness of Facultative
Wetland species
(N_FACW/TOTN_SPP) x 100
c
PCTN_FAC
Percent richness of Facultative
species
(N_FAC/TOTN_SPP) x 100
c
PCTN_FACU
Percent richness of Facultative
Upland species
(N_FACU/TOTN_SPP) x 100
c
PCTN_UPL
Percent richness of UPL (= UPL+ NL)
species
(N_UPL/TOTN_SPP) x 100
c
PCTN_OBL_FACW
Percent richness (number) of
(N_OBL_FACW/TOTN_SPP) x 100
C, in
PRLH-VMMI
Obligate + Facultative Wetland
species
PCTN OBL FACW F
Percent richness (number) of
(N_OBL_FACW_FAC/TOTN_SPP) x
AC
Obligate + Facultative Wetland
species
100
C
PCTN_FAC_FACU
Percent richness (number) of
Facultative + Facultative Upland
species
(N_FAC_FACU/TOTN_SPP) x 100
C
XABCOV_OBL
Mean Absolute Cover of Obligate
S COVER of OBL species across 5
r
species
plots/5 plots
XABCOV_FACW
Mean Absolute Cover of Facultative
S COVER of FACW species across
r
Wetland species
5 plots/5 plots
L
141
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
METRIC NAME
METRIC DESCRIPTION
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
(C = condition,
S = stress)
XABCOV_FAC
Mean Absolute Cover of Facultative
species
S COVER of FAC species across 5
plots/5 plots
C
XABCOV_FACU
Mean Absolute Cover of Facultative
Upland species
S COVER of FACU species across 5
plots/5 plots
C
XABCOV_UPL
Mean Absolute Cover of UPL
species
S COVER of UPL species across 5
plots/5 plots
C
XABCOV
_OBL_FACW
Mean Absolute Cover of Obligate +
Facultative Wetland species
S COVER of OBL and FACW
species across 5 plots/5 plots
C
XABCOV
_OBL_FACW_FAC
Mean Absolute Cover of Obligate +
Facultative Wetland species
S COVER of OBL, FACW, and FAC
species across 5 plots/5 plots
C
XABCOV FAC_FACU
Mean Absolute Cover of Facultative
+ Facultative Upland species
S COVER of FAC and FACU species
across 5 plots/5 plots
C
XRCOV_OBL
Mean Relative Cover of Obligate
species
(XA BCOV_0 B L/XTOTA BCOV) x 100
C
XRCOV_FACW
Mean Relative Cover of Facultative
Wetland species
(XABCOV_FACW/XTOTABCOV) x
100
C
XRCOV_FAC
Mean Relative Cover of Facultative
species
(XABCOV_FAC/XTOTABCOV) x 100
C
XRCOV_FACU
Mean Relative Cover of Facultative
Upland species
(XABCOV_FACU/XTOTABCOV) x
100
C
XRCOV_UPL
Mean Relative Cover of UPL (= UPL)
species
(XA BCO V_U P L/XT OTA BCO V) x 100
C
XRCOV_OBL_FACW
Mean Relative Cover of Obligate +
Facultative Wetland species
(XABCOV _OBL_FACW
/XTOTABCOV) x 100
C
XRCOV_OBL_FACW
_FAC
Mean Relative Cover of Obligate +
Facultative Wetland + Facultative
species
(XABCOV _OBL_FACW_FAC/
XTOTABCOV) x 100
C
XRCOV_FAC_FACU
Mean Relative Cover of Obligate +
Facultative Wetland + Facultative
species
(XABCOV _FAC_FACU/
XTOTABCOV) x 100
C
WETIND_COV_
ALL
Wetland Index, Cover Weighted - all
species
p i p
lij = Importance Value = Mean
absolute cover species / in site j. E,¦ =
Ecological score for species based
on WIS (OBL = 1, FACW = 2, FAC = 3,
FACU = 4, UPL = 5)
X Wl'«
i=l f i= 1
C
WETIND_FRECL
ALL
Wetland Index, Frequency
Weighted - all species
lij = Importance Value = Frequency
for species / in site j. E,¦ = Ecological
score for species based on WIS (OBL
= 1, FACW = 2, FAC = 3, FACU = 4,
UPL = 5)
P j P
w,=i Win
i= 1 ' i= 1
C
142
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
METRIC NAME
WETIND_
COV NAT
METRIC DESCRIPTION
Wetland Index, Cover Weighted -
native species only
lij = Importance Value = Mean
absolute cover for species / in site j.
E, = Ecological score for species
based on WIS (OBL = 1, FACW = 2,
FAC = 3, FACU = 4, UPL = 5)
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
WI
i-1
i— 1
METRIC TYPE
(C = condition,
S = stress)
WETIND_ Wetland Index, Frequency
FREQ_NAT Weighted - native species only
lij = Importance Value = Frequency
for species / in site j. E,¦ = Ecological
score for species based on WIS (OBL
= 1, FACW = 2, FAC = 3, FACU = 4,
UPL = 5)
WI
WETIND2_COV_
ALL
Wetland Index, Cover Weighted - all
species
l,j= Importance Value = Mean WI = ^ Ii} Eij ^ 1^
absolute cover species / in site j. Ei = i=i / i=i
absolute cover species / in site j. £, =
Ecological score for species based
on WIS (OBL = 5, FACW = 4, FAC = 3,
FACU =2, UPL = 1)
WETIND2_FRECL
ALL
Wetland Index, Frequency
Weighted - all species
lij = Importance Value = Frequency
for species / in site j. Ei = Ecological
score for species based on WIS (OBL
= 5, FACW = 4, FAC = 3, FACU =2,
UPL = 1)
f if
M"=XWl'«
£= 1
:=1
WETIND2_
COV NAT
Wetland Index, Cover Weighted -
native species only
lij = Importance Value = Mean
absolute cover for species / in site j.
Ei = Ecological score for species
based on WIS (OBL = 5, FACW = 4,
FAC = 3, FACU =2, UPL= 1)
WI
=zWz,<'
£=1
i= 1
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
143
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
WETIND2_
Wetland Index, Frequency
freclnat
Weighted - native species only
lij = Importance Value = Frequency
p i p
for species / in site j. E,¦ = Ecological
C
score for species based on WIS (OBL
= 5, FACW = 4, FAC = 3, FACU =2,
UPL = 1)
N OBLFACW AC
Number of Alien + Cryptogenic
Count unique ALIEN and CRYP OBL
Obligate and facultative wetland
and FACW species across 100-m2
S
species
plots
XABCOV_
Mean Absolute Cover of Alien +
S COVER of ALIEN and CRYP OBL
OBLFACW_AC
Cryptogenic Obligate and
and FACW species across 5 plots/5
Facultative Wetland species
plots
S
XRCOV
Mean Relative Cover of Alien +
(XA BCO V_0 B LFACW_AC/
OBLFACW_AC
Cryptogenic Obligate and
Facultative Wetland species
XTOTABCOV) x 100
S
SECTION 5
LIFE HISTORY
SECTION 5.1
GROWTH-HABIT
Trait Information = Growth-habit
(Table 5-1); Native Status (Table
5-5)
N_GRAMINOID
Graminoid richness
Count unique GRAMINOID species
across 100-m2 plots
C
N_GRAMINOID_
Native Graminoid richness
Count unique native (NAT)
NAT
GRAMINOID species across 100-
m2 plots
C
N_GRAMINOID_
Alien and cryptogenic Graminoid
Count unique ALIEN and CRYP
AC
richness
GRAMINOID species across 100-
m2 plots
S
N_FORB
Forb richness
Count unique FORB species across
100-m2 plots
C
N_FORB_NAT
Native Forb richness
Count unique native (NAT) FORB
species across 100-m2 plots
C
N_FORB_AC
Alien and cryptogenic Forb richness
Count unique ALIEN and CRYP
FORB species across 100-m2 plots
S
N_HERB
Herbaceous plant (FORB +
GRAMINOID) species richness
N_FORB+ N_GRAMINOID
C
N_HERB_NAT
Native Herbaceous species richness
N_FORB_NAT +
N_GRAMINOID_NAT
C
N_HERB_AC
Alien and cryptogenic Herbaceous
richness
N_FORB_AC + N_GRAMINOID_AC
S
N_SSHRUB_
Subshrub-forb richness
Count unique SUBSHRUB-FORB
r
FORB
species across 100-m2 plots
144
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
N_SSHRUB_
Subshrub-shrub richness
Count unique SUBSHRUB-SHRUB
c
SHRUB
species across 100-m2 plots
N_SHRUB
Shrub richness
Count unique SHRUB species
across 100-m2 plots
c
N SHRUB
Combined Shrub growth-habits
N SHRUB+N SSHRUB SHRUB +
r
COMB
richness
N SSHRUB-FORB
L,
N SHRUB
Native richness of Combined Shrub
Count unique native (NAT)
COMB_NAT
growth-habits richness
SHRUB_COMB species across 100-
m2 plots
c
N_SHRUB_
Alien and cryptogenic richness for
Count unique ALIEN and CRYP
COMB_AC
Combined Shrub growth-habits
SHRUB_COMB species across 100-
m2 plots
s
N_TREE_SHRUB
Tree-Shrub richness
Count unique TREE-SHRUB species
across 100-m2 plots
c
N_TREE
Tree richness
Count unique TREE species across
100-m2 plots
c
N_TREE_COMB
Combined Tree and Tree-Shrub
richness
N_TREE_SHRUB + N_TREE
c
N TREE
Combined Tree and Tree-Shrub
Count unique native (NAT)
COMB_NAT
richness
TREE_COMB species across 100-
m2 plots
c
N TREE
Combined Tree and Tree-Shrub
Count unique ALIEN and CRYP
COMB_AC
richness
TREE_COMB species across 100-
m2 plots
s
N_VINE
Vine richness
Count unique VINE species across
100-m2 plots
c
N_VINE_NAT
Vine richness
Count unique native (NAT) VINE
species across 100-m2 plots
c
N_VINE_AC
Vine richness
Count unique ALIEN and CRYP
VINE species across 100-m2 plots
s
N_VINE_SHRUB
Vine-Shrub richness
Count unique a VINE-SHRUB
species across 100-m2 plots
c
N_VINE_
Native Vine-Shrub richness
Count unique native (NAT) VINE-
SHRUB_NAT
SHRUB species across 100-m2
plots
c
N_VINE_
Alien and cryptogenic Vine-Shrub
Count unique ALIEN and CRYP
SHRUB_AC
richness
VINE-SHRUB species across 100-
m2 plots
s
N_VINE_ALL
Vine-All richness
Count unique a VINE_ALL species
across 100-m2 plots
c
N_VINE_ALL_NAT
Native Vine-All richness
Count unique native (NAT)
VINE_ALL species across 100-m2
plots
c
N_VINE_ALL_AC
Alien and cryptogenic Vine-Shrub
Count unique ALIEN and CRYP
richness
VINE_ALL species across 100-m2
plots
s
PCTN_
Graminoid percent richness
(N_GRAMINOID/TOTN_SPP) x 100
r
GRAMINOID
145
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
PCTN_
Native Graminoid percent richness
(N_GRAMINOID_NAT/TOTN_SPP)
c
GRAMINOID NAT
x 100
PCTN_
Graminoid percent richness
(N_GRAMINOID_AC/TOTN_SPP) x
C
GRAMINOID AC
100
3
PCTN_FORB
Forb percent richness
(N_FORB/TOTN_SPP) x 100
C
PCTN_FORB_
Native Forb percent richness
(N_FORB_NAT/TOTN_SPP) x 100
r
NAT
PCTN_FORB_AC
Alien and cryptogenic Forb percent
richness
(N_FORB_AC/TOTN_SPP) x 100
s
PCTN_HERB
Percent Herbaceous (FORB +
GRAMINOID) richness
(N_HERB/TOTN_SPP) x 100
c
PCTN_HERB_
Percent native Herbaceous richness
(N_HERB_NAT/TOTN_SPP) x 100
r
NAT
L,
PCTN_HERB_
Percent alien and cryptogenic
(N_HERB_AC/TOTN_SPP) x 100
c
AC
Herbaceous richness
j
PCTN_SSHRUB_
Subshrub-Forb percent richness
(N_SSHRUB_FORB/TOTN_SPP) x
r
FORB
100
PCTN_SSHRUB_
Subshrub-Shrub percent richness
(N_SSHRUB/TOTN_SPP) x 100
r
SHRUB
PCTN_SHRUB
Shrub percent richness
(N_S H RU B/TOTN_S P P) x100
c
PCTN SHRUB
Combined Shrub richness
(N_SHRUB_COMB/TOTN_SPP) x
r
COMB
100
L,
PCTN SHRUB
Percent native richness of
(N_SHRUB_COMB_NAT/TOTN_SP
r
COMB_NAT
Combined Shrub growth-habits
P) x 100
L
PCTN_SHRUB_
Percent alien and cryptogenic
(N_SH RU B_COMB_AC/TOTN_SPP)
COMB_AC
richness for Combined Shrub
growth-habits
x 100
s
PCTN_TREE_
Tree-Shrub percent richness
(N_TREE_SHRUB/TOTN_SPP) x
r
SHRUB
100
PCTN_TREE
Tree percent richness
(N_TREE/TOTN_SPP) x 100
C
PCTN_TREE_
Combined Tree and Tree-Shrub
(N_TREE_COMB/TOTN_SPP) x 100
r
COMB
percent richness
PCTN TREE
Combined Tree and Tree-Shrub
(N_TREE_COMB_NAT/TOTN_SPP)
r
COMB_NAT
percent richness
x 100
L,
PCTN_TREE_
Combined Tree and Tree-Shrub
(N_TRE E_CO M B_AC/TOTN_S P P) x
c
COMB_AC
percent richness
100
j
PCTN_VINE
Vine percent richness
(N_VINE/TOTN_SPP) x 100
c
PCTN_VINE_NAT
Native Vine percent richness
(N_VINE_NAT/TOTN_SPP) x 100
c
PCTN_VINE_AC
Alien and cryptogenic Vine percent
richness
(N_VINE_AC/TOTN_SPP) x 100
s
PCTN_VINE_
Vine-Shrub percent richness
(N_VINE_SHRUB/TOTN_SPP) x 100
r
SHRUB
PCTN_VINE_
Native Vine-Shrub percent richness
(N_VINE_SHRUB_NAT/TOTN_SPP)
r
SHRUB_NAT
x 100
PCTN_VINE_
Alien and Cryptogenic Vine-Shrub
(N_VI N E_S H RU B_AC/TOTN_S P P) x
c
SHRUB_AC
percent richness
100
3
146
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
PCTN_VINE_ALL
All-Vine percent richness
(N_VINE_ALL/TOTN_SPP) x 100
C
PCTN_VINE_ALL_NA
All-Vine native percent richness
(N_VINE_ALL_NAT/TOTN_SPP) x
r
T
100
P CT N_V 1N E_A L L_AC
All-Vine alien and cryptogenic
(N_VINE_ALL_AC/TOTN_SPP) x
c
percent richness
100
o
XABCOV_
Mean absolute Graminoid cover
S COVER of GRAMINOID species
r
GRAMINOID
across 5 plots/5 plots
L,
XABCOV_
Mean absolute native Graminoid
S COVER of GRAMINOID NAT
r
GRAMINOID_NAT
cover
species across 5 plots/5 plots
L,
XABCOV_
Mean absolute alien and
S COVER of GRAMINOID ALIEN
GRAMINOID_AC
cryptogenic Graminoid cover
and CRYP species across 5 plots/5
plots
s
XABCOV_FORB
Mean absolute FORB cover
S COVER of FORB species across 5
plots/5 plots
c
XABCOV_FORB_
Mean absolute native FORB cover
S COVER of NAT FORB species
r
NAT
across 5 plots/5 plots
L,
XABCOV_FORB_
Mean absolute alien and
S COVER of ALIEN and CRYP FORB
c
AC
cryptogenic FORB cover
species across 5 plots/5 plots
j
XABCOV HERB
Mean absolute Herbaceous species
XABCOV FORB +
r
cover (FORB + GRAMINOID)
XABCOV GRAMINOID
L,
XABCOV HERB
Mean absolute native Herbaceous
XABCOV FORB NAT +
r
NAT
cover
XABCOV GRAMINOID NAT
L,
XABCOV HERB
Mean relative Herbaceous alien and
XABCOV FORB AC +
c
AC
cryptogenic cover
XABCOV_GRAMINOID_AC
o
XABCOV_
Mean absolute Subshrub-Forb
S COVER of SUBSHRUB-FORB
r
SSHRUB_FORB
cover
species across 5 plots/5 plots
L
XABCOV_
Mean absolute Subshrub-Shrub
S COVER SUBSHRUB-SHRUB
r
SSHRUB_SHRUB
cover
species across 5 plots/5 plots
L
XABCOV_SHRUB
Mean absolute Shrub cover
S COVER of SHRUB species across
5 plots/5 plots
c
XABCOV_
Combined Shrub growth-habits
S COVER of SHRUB_COMB species
r
SHRUB_COMB
absolute cover
across 5 plots/5 plots
L,
XABCOV SHRUB
Mean absolute native Combined
S COVER of NAT SHRUB-COMB
r
COMB_NAT
Shrub growth-habits cover
species across 5 plots/5 plots
L
XABCOV_SHRUB_
Mean absolute alien and
S COVER of ALIEN and CRYP
COMB_AC
cryptogenic Combined Shrub
growth-habits cover
SHRUB_COMB species across 5
plots/5 plots
s
XABCOV_TREE_
Mean absolute Tree-Shrub cover
S COVER of TREE-SHRUB species
r
SHRUB
across 5 plots/5 plots
L,
XABCOV_TREE
Mean absolute Tree cover
S COVER of TREE species across 5
plots/5 plots
c
XABCOV_TREE_
Combined Tree and Tree-Shrub
S COVER of TREE_COMB species
r
COMB
absolute cover
across 5 plots/5 plots
L
XABCOV_TREE_
Combined native Tree and Tree-
S COVER of NAT TREE_COMB
r
COMB_NAT
Shrub absolute cover
species across 5 plots/5 plots
L
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
XABCOV_TREE_
Combined alien and cryptogenic
S COVER of ALIEN and CRYP
COMB_AC
Tree and Tree-Shrub absolute cover
TREE_COMB species across 5
plots/5 plots
S
XABCOV_VINE
Mean absolute Vine cover
S COVER of VINE species across 5
plots/5 plots
C
XABCOV_VINE_
Mean native absolute Vine cover
S COVER of NAT VINE species
c
NAT
across 5 plots/5 plots
L
XABCOV_VINE_
Mean alien and cryptogenic
S COVER of ALIEN and CRYP VINE
c
AC
absolute Vine cover
species across 5 plots/5 plots
o
XABCOV_VINE_
Mean absolute Vine-Shrub cover
S COVER of VINE-SHRUB species
c
SHRUB
across 5 plots/5 plots
L
XABCOV_VINE_
Mean absolute native Vine-Shrub
S COVER of NAT VINE-SHRUB
c
SHRUB_NAT
cover
species across 5 plots/5 plots
L
XABCOV_VINE_
Mean absolute alien and
S COVER of ALIEN and CRYP VINE-
SHRUB_AC
cryptogenic Vine-Shrub cover
SHRUB species across 5 plots/5
plots
s
XABCOV_VINE_
Mean absolute Vine-ALL cover
S COVER of VINE-ALL species
c
ALL
across 5 plots/5 plots
L
XABCOV_VINE_
Mean absolute native Vine-ALL
S COVER of NAT VINE-ALL species
c
ALL_NAT
cover
across 5 plots/5 plots
L
XABCOV_VINE_
Mean absolute alien and
S COVER of ALIEN and CRYP VINE-
c
ALL_AC
cryptogenic Vine-ALL cover
ALL species across 5 plots/5 plots
o
XRCOV_
Mean relative Graminoid cover
(XABCOV_GRAMINOID/
C,
GRAMINOID
XTOTABCOV) x 100
in EW-VMMI
XRCOV
Mean relative native Graminoid
(XABCOV_GRAMINOID_NAT/
c
GRAMINOID NAT
cover
XTOTABCOV) x 100
XRCOV_
Mean relative alien and cryptogenic
(XABCOV_G RAM 1 NO 1 D_AC/
C
GRAMINOID AC
Graminoid cover
XTOTABCOV) x 100
J
XRCOV_FORB
Mean relative Forb cover
(XABCOV_FORB/XTOTABCOV) x
100
C,
in EH-VMMI
XRCOV
Mean relative native Forb cover
(XABCOV_FORB_NAT/
c
FORB_NAT
XTOTABCOV) x 100
L
XRCOV_FORB_AC
Mean relative alien and cryptogenic
(XA BCO V_FO R B_AC/XT OT A BCOV)
r
Forb cover
x 100
XRCOV_HERB
Mean relative Herbaceous (FORB +
(XABCOV_HERB/XTOTABCOV) x
r
GRAMINOID) cover
100
XRCOV
Mean relative native Herbaceous
(XABCOV_HERB_NAT/
c
HERB NAT
cover
XTOTABCOV)x100
XRCOV_HERB_AC
Mean relative alien and cryptogenic
(XABCOV_HERB_AC/XTOTABCOV)
C
Herbaceous cover
x 100
3
XRCOV_SSHRUB_
Mean relative Subshrub-Forb cover
(XABCOV_SSHRUB_FORB/
c
FORB
XTOTABCOV) x 100
L
XRCOV_SSHRUB_
Mean relative Subshrub-Shrub
(XA BCO V_SS H RU B_S H RU B/
c
SHRUB
cover
XTOTABCOV) x 100
L
XRCOV_SHRUB
Mean relative Shrub cover
(XABCOV_SHRUB/XTOTABCOV) x
100
c
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
XRCOV SHRUB
Mean relative Combined Shrub
(XABCOV_S H RU B_COMB/
r
COMB
growth-habits cover
XTOTABCOV) x 100
L
XRCOV SHRUB
Mean relative native Combined
(XABCOV_S H RU B_COM B_N AT/
r
COMB NAT
Shrub growth-habits cover
XTOTABCOV) x 100
L,
XRCOV_SHRUB_
Mean relative alien and cryptogenic
(XABCOV_S H RU B_COM B_AC/
COMB_AC
Combined Shrub growth-habits
cover
XTOTABCOV) x 100
s
XRCOV TREE
Mean relative Tree-Shrub cover
(XABCOV_TREE_SHRUB/
r
SHRUB
XTOTABCOV) x 100
L
XRCOV_TREE
Mean relative Tree cover
(XABCOV_TREE/XTOTABCOV) x
100
c
XRCOV TREE
Mean relative Combined Tree and
(XA BCO V_TR E E_CO MB/
r
COMB
Tree-Shrub cover
XTOTABCOV) x 100
L,
XRCOV TREE
Mean relative Combined Tree and
(XABCOV_TRE E_CO M B_N AT/
r
COMB NAT
Tree-Shrub cover
XTOTABCOV) x 100
L,
XRCOV TREE
Mean relative Combined Tree and
(XA BCO V_TR E E_CO M B_AC/
c
COMB AC
Tree-Shrub cover
XTOTABCOV) x 100
j
XRCOV_VINE
Mean relative Vine cover
(XABCOV_VINE/XTOTABCOV) x
100
c
XRCOV_VINE_
Mean native relative Vine cover
(XABCOV_VIN E_NAT/XTOTABCOV)
r
NAT
x 100
XRCOV_VINE_
Mean alien and cryptogenic relative
(XABCOV_VIN E_AC/XTOTABCOV)
c
AC
Vine cover
x 100
o
XRCOV VINE
Mean relative Vine-Shrub cover
(XA BCO V_V 1N E_S H RU B/
r
SHRUB
XTOTABCOV) x 100
L,
XRCOV VINE
Mean native relative Vine-Shrub
(XABCOV_VIN E_SH RU B_NAT/
c
SHRUB NAT
cover
XTOTABCOV) x 100
XRCOV_VINE_
Mean alien and cryptogenic relative
(XABCOV_VIN E_SH RU B_AC/
C
SHRUB_AC
Vine-Shrub cover
XTOTABCOV) x 100
J
XRCOV VINE
Mean relative Vine-ALL cover
(XABCOV_VINE_ALL/
r
ALL
XTOTABCOV) x 100
L
XRCOV VINE
Mean native relative Vine-ALL cover
(XABCOV_VIN E_ALL_NAT/
r
ALL NAT
XTOTABCOV) x 100
L
XRCOV_VINE_
Mean alien and cryptogenic relative
(XABCOV_VIN E_ALL_AC/
c
ALL AC
Vine-ALL cover
XTOTABCOV) x 100
j
Section 5.2
DURATION
Trait Information = Duration
(Table 5-2); Native Status (Table
5-5)
N_ANNUAL
Annual species richness
Count unique ANNUAL species
across 100-m2 plots
c,
In EH-VMMI
N_ANNUAL_NAT
Native Annual richness
Count unique NAT ANNUAL
species across 100-m2 plots
C
N_ANNUAL_AC
Alien and cryptogenic Annual
Count unique ALIEN and CRYP
richness
ANNUAL species across 100-m2
plots
S
N_ANN_BIEN
Annual-Biennial richness
Count unique ANN_BIEN species
across 100-m2 plots
c
149
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
N_ANN_
Native Annual-Biennial richness
Count unique NAT ANN_BIEN
c
BIEN NAT
species across 100-m2 plots
N_ANN_
Alien and cryptogenic Annual-
Count unique ALIEN and CRYP
BIEN_AC
Biennial richness
ANN_BIEN species across 100-m2
plots
s
N_ANN_PEREN
Annual-Perennial richness
Count unique ANN_PEREN species
across 100-m2 plots
c
N_ANN_
Native Annual-Perennial richness
Count unique NAT ANN_PEREN
r
PEREN_NAT
species across 100-m2 plots
L,
N_ANN_
Alien and cryptogenic Annual-
Count unique ALIEN and CRYP
PEREN_AC
Perennial richness
ANN_PEREN species across 100-
m2 plots
s
N_PERENNIAL
Perennial richness
Count unique PERENNIAL species
across 100-m2 plots
c
N_PERENNIAL_
Native Perennial richness
Count unique NAT PERENNIAL
r
NAT
species across 100-m2 plots
N PERENNIAL AC
Alien and cryptogenic Perennial
Count unique ALIEN and CRYP
richness
PERENNIAL species across 100-m2
plots
S
PCTN_ANNUAL
Percent Annual richness
(N_ANNUAL/TOTN_SPP) x 100
C
PCTN ANNUAL
Percent native Annual richness
(N_ANNUAL_NAT/TOTN_SPP) x
c
NAT
100
L
PCTN_ANNUAL_
Percent alien and cryptogenic
(N_ANNUAL_AC/TOTN_SPP) x 100
c
AC
Annual richness
j
PCTN_ANN_BIEN
Percent Annual-Biennial richness
(N_ANN_BIEN/TOTN_SPP) x 100
c
PCTN ANN
Percent native Annual-Biennial
(N_ANN_BIEN_NAT/TOTN_SPP) x
r
BIEN NAT
richness
100
L,
PCTN_ANN_
Percent alien and cryptogenic
(N_ANN_BIEN_AC/TOTN_SPP) x
C
BIEN AC
Annual-Biennial richness
100
3
PCTN ANN
Percent Annual-Perennial richness
(N_ANN_PEREN/TOTN_SPP) x 100
r
PEREN
L,
PCTN ANN
Percent native Annual-Perennial
(N_ANN_PEREN_NAT/TOTN_SPP)
c
PEREN_NAT
richness
x 100
L
PCTN_ANN_
Percent alien and cryptogenic
(N_ANN_PEREN_AC/TOTN_SPP) x
C
PEREN AC
Annual-Perennial richness
100
j
PCTN_PERENNIAL
Percent Perennial richness
(N_PERENNIAL/TOTN_SPP) x 100
c
PCTN
Percent native Perennial richness
(N_PERENNIAL_NAT/TOTN_SPP) x
c
PERENNIAL NAT
100
L
PCTN_
Percent alien and cryptogenic
(N_PERENNIAL_AC/TOTN_SPP) x
c
PERENNIAL AC
Perennial richness
100
3
XABCOV_
Mean absolute Annual cover
S COVER of ANNUAL species
r
ANNUAL
across 5 plots/5 plots
L,
XABCOV_
Mean absolute native Annual cover
S COVER of NAT ANNUAL species
r
ANNUAL_NAT
across 5 plots/5 plots
L,
XABCOV_
Mean absolute alien and
S COVER of ALIEN and CRYP
ANNUAL_AC
cryptogenic Annual cover
ANNUAL species across 5 plots/5
plots
s
150
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
XABCOV_ANN_
Mean absolute Annual-Biennial
S COVER of ANN_BIEN species
r
BIEN
cover
across 5 plots/5 plots
L
XABCOV_ANN_
Mean absolute native Annual-
S COVER of NAT ANN BIEN
r
BIEN_NAT
Biennial cover
species across 5 plots/5 plots
L
XABCOV_ANN_
Mean absolute alien and
S COVER of ALIEN and CRYP
BIEN_AC
cryptogenic Annual-Biennial cover
ANN_BIEN species across 5 plots/5
plots
s
XABCOV_ANN_
Mean absolute Annual-Perennial
S COVER of ANN_PEREN species
r
PEREN
cover
across 5 plots/5 plots
L
XABCOV_ANN_
Mean absolute native Annual-
S COVER of NAT ANN PEREN
r
PEREN_NAT
Perennial cover
species across 5 plots/5 plots
L
XABCOV_ANN_
Mean absolute alien and
S COVER of ALIEN and CRYP
PEREN_AC
cryptogenic Annual-Perennial cover
ANN_PEREN species across 5
plots/5 plots
s
XABCOV_
Mean absolute Perennial cover
S COVER of PERENNIAL species
r
PERENNIAL
across 5 plots/5 plots
L
XABCOV_
Mean absolute native Perennial
S COVER of NAT PERENNIAL
r
PERENNIAL_NAT
cover
species across 5 plots/5 plots
L
XABCOV_
Mean absolute alien and
S COVER of ALIEN and CRYP
PERENNIAL_AC
cryptogenic Perennial cover
PERENNIAL species across 5
plots/5 plots
s
XRCOV_ANNUAL
Mean relative annual cover
(XABCOV_ANNUAL/XTOTABCOV) x
100
c
XRCOV ANNUAL
Mean relative native Annual cover
(XABCOV_ANNUAL_NAT/
r
NAT
XTOTABCOV) x 100
L,
XRCOV ANNUAL
Mean relative alien and cryptogenic
(XABCOV_ANNUAL_AC/
c
AC
Annual cover
XTOTABCOV) x 100
j
XRCOV ANN
Mean relative Annual-Biennial
(XA BCO V_A N N_B 1E N/
r
BIEN
cover
XTOTABCOV) x 100
L
XRCOV ANN
Mean relative native Annual-
(XABCOV_ANN_BIEN_NAT/
r
BIEN NAT
Biennial cover
XTOTABCOV) x 100
L
XRCOV_ANN_
Mean relative alien and cryptogenic
(XABCOV_ANN_BIEN_AC/
c
BIEN AC
Annual-Biennial cover
XTOTABCOV) x 100
j
XRCOV ANN
Mean relative Annual-Perennial
(XABCOV_ANN_PEREN/
r
PEREN
cover
XTOTABCOV)x100
L,
XRCOV ANN
Mean relative native Annual-
(XABCOV_ANN_PEREN_NAT/
r
PEREN NAT
Perennial cover
XTOTABCOV)x100
L,
XRCOV_ANN_
Mean relative alien and cryptogenic
(XA BCOV_A N N_P E R E N_AC/
c
PEREN_AC
Annual-Perennial cover
XTOTABCOV)x100
j
XRCOV_
Mean relative Perennial cover
(XABCOV_PERENNIAL/
r
PERENNIAL
XTOTABCOV) x 100
L,
XRCOV_
Mean relative native Perennial
(XABCOV_PERENNIAL_NAT/
r
PERENNIAL_NAT
cover
XTOTABCOV) x 100
L,
XRCOV_
Mean relative alien and cryptogenic
(XABCOV_PERENNIAL_AC/
C
PERENNIAL_AC
Perennial cover
XTOTABCOV) x 100
3
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
Section 5.3
PLANT CATEGORY
Trait Information = Plant
Category (See Section 5.6.3);
Native Status (Table 5-5)
N_DICOT
Dicot richness
Count unique DICOT species
across 100-m2 plots
C
N_DICOTS_NAT
Native Dicot richness
Count unique NAT DICOT species
across 100-m2 plots
C
N_DICOTS_AUEN
Alien Dicot richness
Count unique ALIEN DICOT species
across 100-m2 plots
S
N_DICOTS_CRYP
Cryptogenic Dicot richness
Count unique CRYP DICOT species
across 100-m2 plots
C
N_DICOTS_AC
Alien and Cryptogenic richness
N_DICOT_ALIEN + N_DICOT_CRYP
S
N_FERN
Fern richness
Count unique FERN species across
100-m2 plots
C
N_FERNS_NAT
Native Fern richness
Count unique native FERN species
across 100-m2 plots
C
N_FERNS_INTR
Introduced FERN species richness
Count unique introduced FERN
species across 100-m2 plots
S
N_GYMNOSPERM
Gymnosperm richness
Count unique GYMNOSPERM
species across 100-m2 plots
C
N_LYCOPOD
Lycopod richness
Count unique LYCOPOD species
across 100-m2 plots
C
N_HORSETAIL
Horsetail richness
Count unique HORSETAIL species
across 100-m2 plots
C
N_MONOCOT
Monocot richness
Count unique MONOCOT species
across 100-m2 plots
C
N_MONOCOTS_
Native Monocot richness
Count unique NAT MONOCOT
r
NAT
species across 100-m2 plots
N MONOCOTS
Alien Monocot richness
Count unique ALIEN MONOCOT
c
ALIEN
species across 100-m2 plots
j
N_MONOCOTS_
Cryptogenic Monocot richness
Count unique CRYP MONOCOT
c
CRYP
species across 100-m2 plots
J
N MONOCOTS
Alien and cryptogenic Monocot
N MONOCOT ALIEN +
c
AC
richness
N MONOCOT CRYP
o
PCTN_DICOT
Dicot percent richness
(N_DICOTS/TOTN_SPP) x 100
c
PCTN_DICOTS_
Native Dicot percent richness
(N_DICOTS_NAT/TOTN_SPP) x 100
r
NAT
PCTN_DICOTS_
Alien Dicot percent richness
(N_DICOTS_ALIEN/TOTN_SPP) x
c
ALIEN
100
PCTN_DICOTS_
Cryptogenic Dicot percent richness
(N_DICOTS_CRYP/TOTN_SPP) x
c
CRYP
100
PCTN_DICOTS_AC
Alien and cryptogenic Dicot percent
richness
(N_DICOTS_AC/TOTN_SPP) x 100
S
PCTN_FERN
Fern percent richness
(N_FERNS/TOTN_SPP) x100
C
PCTN_FERNS_
Native Ferns percent richness
(N_FERNS_NAT/TOTN_SPP) x 100
r
NAT
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
PCTN_FERNS_
Introduced Fern percent richness
(N_FERNS_INTR/TOTN_SPP) x 100
C
INTR
J
PCTN_
GYMNOSPERM Percent Richness
(N_GYNOSPERM/TOTN_SPP) x 100
r
GYMNOSPERM
PCTN_LYCOPOD
Lycopod percent richness
(N_LYCOPOD/TOTN_SPP) x 100
c
PCTN_HORSETAIL
Horsetail percent richness
(N_HORSETAIL/TOTN_SPP) x 100
c
PCTN_
Monocot percent richness
(N_MONOCOTS/TOTN_SPP) x 100
C,
MONOCOT
in EW-VMMI
PCTN_
Native Monocot percent richness
(N_MONOCOTS_NAT/TOTN_SPP)
r
MONOCOTS NAT
x 100
PCTN
Alien Monocot percent richness
(N_MONOCOTS_ALIEN/
MONOCOTS
TOTN_SPP) x 100
s
ALIEN
PCTN_
Cryptogenic Monocot percent
(N_MONOCOTS_CRYP/TOTN_SPP)
MONOCOTS
richness
x 100
s
CRYP
PCTN_
Alien and cryptogenic monocot
(N_MONOCOTS_AC/TOTN_SPP) x
c
MONOCOTS AC
percent richness
100
XABCOV_DICOT
Mean absolute cover Dicots
S COVER of DICOT species across
5 plots/5 plots
c
XABCOV_
Mean absolute cover native Dicots
S COVER of NAT DICOT species
r
DICOTS_NAT
across 5 plots/5 plots
L,
XABCOV_
Mean absolute cover Alien Dicots
S COVER of ALIEN DICOT species
c
DICOTS_ALIEN
across 5 plots/5 plots
j
XABCOV_
Mean absolute cover cryptogenic
S COVER of CRYP DICOT species
c
DICOTS_CRYP
Dicots
across 5 plots/5 plots
J
XABCOV
Mean absolute cover of alien and
XABCOV DICOTS ALIEN +
c
DICOTS AC
cryptogenic Dicots
XABCOV DICOTS CRYP
J
XABCOV_FERN
Mean absolute cover of Ferns
S COVER of FERN species across 5
plots/5 plots
c
XABCOV_FERNS_
Mean absolute cover of native
S COVER of NAT FERN species
r
NAT
Ferns
across 5 plots/5 plots
L,
XABCOV_FERNS_
Mean absolute cover of introduced
S COVER o ntroduced INTR FERN
c
INTR
Ferns
species across 5 plots/5 plots
j
XABCOV_
Mean absolute cover of
S COVER of GYMNOSPERM
r
GYMNOSPERM
Gymnosperms
species across 5 plots/5 plots
L,
XABCOV_
Mean absolute cover of Lycopods
S COVER of LYCOPOD species
r
LYCOPOD
across 5 plots/5 plots
XABCOV_
Mean absolute cover of Horsetails
S COVER of HORSETAIL species
r
HORSETAIL
across 5 plots/5 plots
L,
XABCOV_
Mean absolute cover of Monocots
S COVER of MONOCOT species
r
MONOCOT
across 5 plots/5 plots
L,
XABCOV_
Mean absolute cover of native
S COVER of NAT MONOCOT
c
MONOCOTS_NAT
Monocots
species across 5 plots/5 plots
XABCOV_
Mean absolute cover of alien
S COVER of ALIEN MONOCOT
MONOCOTS_
Monocots
species across 5 plots/5 plots
s
ALIEN
153
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
XABCOV_
Mean absolute cover of cryptogenic
S COVER of CRYP MONOCOT
MONOCOTS_
Monocots
species across 5 plots/5 plots
S
CRYP
XABCOV
Mean absolute cover of alien and
XABCOV MONOCOTS ALIEN +
C
MONOCOTS AC
cryptogenic Monocots
XABCOV MONOCOTS CRYP
J
XRCOV_DICOT
Mean relative cover Dicots
(XABCOV_DICOTS/XTOTABCOV) x
100
c
XRCOV DICOTS
Mean relative cover native Dicots
(XABCOV_DICOTS_NAT/
c
NAT
XTOTABCOV) x 100
L
XRCOV DICOTS
Mean relative cover alien Dicots
(XABCOV_DICOTS_ALIEN/
c
ALIEN
XTOTABCOV) x 100
o
XRCOV_DICOTS_
Mean relative cover cryptogenic
(XA BCO V_D 1 COTS_C RYP/
c
CRYP
Dicots
XTOTABCOV) x 100
3
XRCOV DICOTS
Mean relative cover of alien and
(XA BCO V_D 1 COTS_AC/
c
AC
cryptogenic Dicots
XTOTABCOV) x 100
J
XRCOV_FERN
Mean relative cover of Ferns
(XABCOV_FERNS/
XTOTABCOV) x 100
c
XRCOV FERNS
Mean relative cover of native Ferns
(XABCOV_FERNS_NAT/
c
NAT
XTOTABCOV) x 100
L
XRCOV FERNS
Mean relative cover of introduced
(XABCOV_F E RNS_I NTR/
c
INTR
Ferns
XTOTABCOV) x 100
o
XRCOV_
Mean relative cover of
(XABCOV_GYMNOSPERMS/
r
GYMNOSPERM
Gymnosperms
XTOTABCOV) x 100
XRCOV_LYCOPOD
Mean relative cover of Lycopods
(XABCOV_LYCOPODS/
XTOTABCOV) x 100
c
XRCOV_
Mean relative cover of Horsetails
(XABCOV_HORSETAILS/
r
HORSETAIL
XTOTABCOV) x 100
XRCOV_
Mean relative cover of Monocots
(XABCOV_MONOCOTS/
r
MONOCOT
XTOTABCOV) x 100
XRCOV
Mean relative cover of native
(XABCOV_MONOCOTS_NAT/
C, in EH-
MONOCOTS NAT
Monocots
XTOTABCOV) x 100
VMMI, PRLW-
¦,
VMMI, 2011
National
VMMI
XRCOV
Mean relative cover of alien
(XA BCO V_M O N OCOTS_A LI E N/
MONOCOTS_
Monocots
XTOTABCOV) x 100
S
ALIEN
XRCOV_
Mean relative cover of cryptogenic
(XA BCO V_M O N OCOTS_C RYP/
MONOCOTS_
Monocots
XTOTABCOV) x 100
S
CRYP
XRCOV_
Mean relative cover of alien and
(XA BCO V_M O N OCOTS_AC/
c
MONOCOTS_AC
cryptogenic Monocots
XTOTABCOV) x 100
o
Sections 6 - 8 METRICS BASED ON FIELD DATA FROM FORM V-l: NWCA 2016
VEGETATION PLOT ESTABLISHMENT AND FORM V-3: NWCA 2016
VEGETATION TYPES (FRONT) AND NWCA 2016 GROUND SURFACE
ATTRIBUTES (BACK)
154
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
SECTION 6
WETLAND TYPE HETEROGENEITY
BASED ON PLOT-LEVEL NWCA
WETLAND TYPES (designated as
'Predominant NWCA Wetland
Type' on Form V-l)
N SANDT
Number of unique NWCA Wetland
Count number of unique NWCA
c
Types ( l/ETLAND_TYPE) in AA
WETLAND_TYP across the 5 plots
DOM_SANDT
Dominant NWCA
WETLAND_TYPI (s) in AA
Select dominant NWCA
WETLAND_TYPE: Most frequent
(greatest number of plots), or in
case of ties, the two most
frequent hyphenated
c
D_SANDT
Simpson's Diversity - Heterogeneity
of NWCA A/ETLAND_TYPI s in AA
s = number of S&T classes present, /
= class i, p = proportion of S&T
Classes belonging to class /
D = l —
i
c
H_SANDT
Shannon-Wiener - Heterogeneity of
NWCA i/ETLAND_TYPE s in AA
s = number of S&T classes present, /
= class i, p = proportion of S&T
Classes belonging to class /
s
H' =
i
c
J_SANDT
Pielou Evenness - Heterogeneity of
H'
' ~ hiS
NWCA i/ETLAND_TYPE s in AA
c
S = number of S&T classes observed
SECTION 7
VEGETATION STRUCTURE/TYPES
SECTION 7.1
Vascular Strata
N_VASC_STRATA
Number of unique Vascular
Count number of unique vascular
Vegetation Strata across AA
vegetation strata across the 5
plots
c
XN_VASC_
Mean number of vascular
r
STRATA
vegetation strata across plots
RG VASC
Range in number of vascular
Maximum - minimum number of
STRATA
vegetation strata found in all 100-
m2 plots
vegetation strata across five 100-
m2 plots
c
XTOTCOV_VASC_
Mean total cover of all vascular
(S cover for all vascular strata
r
STRATA
strata
across all 100-m2 plots)/5 plots
L
FRECL
Frequency Submerged Aquatic
(# of 100-m2 plots in which
SUBMERGED_AQ
Vegetation
SUBMERGED_AQoccurs/5 plots) x
100
c
FREQ_FLOATING_
Frequency Floating Aquatic
(# of 100-m2 plots in which
AQ
Vegetation
FLOATING_AQ occurs/5 plots) x
100
c
155
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
FRECLLIANAS
Frequency Lianas, vines, and
(# of 100-m2 plots in which
r
vascular epiphytes
LIANAS occi s/5 plots) x 100
FRECLVTALL_VEG
Frequency Vegetation > 30m tall
(# of 100-m2 plots in which
VTALL_VEG occurs/5 plots) x 100
C
FREQ_TALL_VEG
Frequency Vegetation > 15m to
(# of 100-m2 plots in which
r
30m tall
TALL_VEG occurs/5 plots) x 100
freclhmed_
Frequency Vegetation > 5m to 15m
(# of 100-m2 plots in which
r
VEG
tall
HMED_VEG occ rs/5 plots) x 100
FREQ_MED_VEG
Frequency Vegetation >2m to 5 tall
(# of 100-m2 plots in which
MED_VEG occurs/5 plots) x 100
C
FREQ_SMALL_
Frequency Vegetation 0.5 to 2m tall
(# of 100-m2 plots in which
r
VEG
SMALL_VEG occurs/5 plots) x 100
FRECLVSMALL_
Frequency Vegetation < 0.5m tall
(# of 100-m2 plots in which
VEG
VSMALL_VEG occ rs/5 plots) x
100
C
XCOV_
Mean absolute cover Submerged
S cover of UBMERGED_AQ
r
SUBMERGED_AQ
Aquatic Vegetation
across 5 plots/5 plots
XCOV_
Mean absolute cover Floating
S cover of LOATING AC across 5
r
FLOATING_AQ
Aquatic Vegetation
plots/5 plots
L,
XCOV_LIANAS
Mean absolute cover Lianas, vines,
S cover of IANA! across 5 plots/5
r
and vascular epiphytes
plots
L,
XCOV_VTALL_
Mean absolute cover Vegetation >
S cover of 'TALL VE< across 5
r
VEG
30m tall
plots/5 plots
L,
XCOV_TALL_VEG
Mean absolute cover Vegetation >
S cover of ALL VEG across 5
r
15m to 30m tall
plots/5 plots
L,
XCOV_HMED_
Mean absolute cover Vegetation >
S cover of 1MED VEG across 5
r
VEG
5m to 15m tall
plots/5 plots
L,
XCOV_MED_VEG
Mean absolute cover Vegetation
S cover of /IED VEG across 5
r
>2m to 5 tall
plots/5 plots
L,
XCOV_SMALL_
Mean absolute cover Vegetation 0.5
S cover of MALL VEG across 5
r
VEG
to 2m tall
plots/5 plots
L,
XCOV VSMALL
Mean absolute cover Vegetation <
Jcover of VSMALL_VE( across 5
r
VEG
0.5m tall
plots/5 plots
L,
IMP_
Importance Submerged Aquatic
(FRECLSUBMERGED_AQ +
r
SUBMERGED_AQ
Vegetation
XCO V_S U B M E RG E D_AQ)/2
IMP_FLOATING_
Importance Floating Aquatic
(freclfloating_aq +
r
AQ
Vegetation
XCO V_F LO ATI N G_AQ)/2
IMP_LIANAS
Importance Lianas, vines, and
vascular epiphytes
(FRECLLIANAS + XCOV_LIANAS)/2
C
IMP_VTALL_VEG
Importance Vegetation > 30m tall
(FRECLVTALL_VEG +
XCOV_VTALL_VEG)/2
C
IMP_TALL_VEG
Importance Vegetation > 15m to
(FREQJ"ALL_VEG +
r
30m tall
XCOV_TALL_VEG)/2
IMP_HMED_VEG
Importance Vegetation > 5m to
(freclhmed_veg +
r
15m tall
XCOV_HMED_VEG )/2
IMP_MED_VEG
Importance Vegetation >2m to 5 tall
(FREQJVIED_VEG +
XCOV_MED_VEG)/2
C
156
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
IMP_SMALL_VEG
Importance Vegetation 0.5 to 2m
(FRECLSMALL_VEG +
r
tall
XCOV_SMALL_VEG)/2
IMP_VSMALL_
Importance Vegetation < 0.5m tall
(FRECLVSMALL_VEG +
r
VEG
XCOV_VSMALL_VEG)/2
XRCOV_
Relative mean cover Submerged
(XCOV_S U BM E RG E D_AQ/
r
SUBMERGED_AQ
Aquatic Vegetation
XTOTCOV_VASC_STRATA) x 100
XRCOV_
Relative mean cover Floating
(XCOV_F LO AT 1N G_AQ/
r
FLOATING_AQ
Aquatic Vegetation
XTOTCOV_VASC_STRATA) x 100
XRCOV LIANAS
Relative cover Lianas, Vines, and
(XCOV_LIANAS/
c
Vascular Epiphytes
XTOTCOV_VASC_STRATA) x 100
L
XRCOV_VTALL_
Relative cover Vegetation > 30m tall
(XCOV_VTALL_VEG/
r
VEG
XTOTCOV_VASC_STRATA) x 100
L,
XRCOV_TALL_
Relative cover Vegetation > 15m to
(XCOV_TALL_VEG/
r
VEG
30m tall
XTOTCOV_VASC_STRATA) x 100
XRCOV_HMED_
Relative cover Vegetation > 5m to
(XCOV_HMED_VEG/
r
VEG
15m tall
XTOTCOV_VASC_STRATA) x 100
XRCOV_MED_
Relative cover Vegetation >2m to 5
(XCOV_MED_VEG/
r
VEG
tall
XTOTCOV_VASC_STRATA) x 100
XRCOV_SMALL_
Relative cover Vegetation 0.5 to 2m
(XCOV_S MA LL_V EG/
r
VEG
tall
XTOTCOV_VASC_STRATA) x 100
L,
XRCOV_VSMALL_
Relative cover Vegetation < 0.5m
(XCOV_VSMALLJ
r
VEG
tall
XTOTCOV_VASC_STRATA) x 100
D_VASC_STRATA
Simpson's Diversity - Heterogeneity
of Vertical Vascular Structure in AA
based on occurrence and relative
cover of all strata in all plots
s = number of veg strata observed, /
= veg stratum /', p = relative cover
belonging to veg stratum /
3
D =
I
c
H_VASC_STRATA
Shannon-Wiener - Heterogeneity of
Vertical Vascular Structure in AA
based on occurrence and relative
3
H' = ~ V Pi In Pi
cover of all strata in all plots
s '
I
c
s = number of veg strata observed, /
= veg stratum /', p = relative cover
belonging to veg stratum /
J_VASC_STRATA
Pielou Evenness - Heterogeneity of
Vertical Vascular Structure in AA
H'
¦! ~\n5
based on occurrence and relative
r
cover of all strata in all plots
5=number of strata observed
Section 7.2 Non-Vascular Groups
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric METRIC TYpE
l\UVV/; . .
„ 1T .. . ,. (C = condition.
SPECIES TRAIT TYPE (indicated in
METRIC NAME METRIC DESCRIPTION Banner if applicable) -stress)
N_PEAT_MOSS_
DOM
Number of plots where bryophytes
are dominated by Sphagnum or
other peat forming moss
Count number of plots where
PEAT_MOSS = Y
C
FREQ_PEAT_
MOSS_DOM
Frequency of plots where
bryophytes are dominated by
Sphagnum or other peat forming
moss
(N_PEAT_MOSS_DOM/5 plots) x
100
C
FRECL
BRYOPHYTES
Frequency of bryophytes growing
on ground surfaces, logs, rocks, etc.
(# of 100-m2 plots in which
BRYOPHYTES occur/5 plots) x 100
C
FREQJ-ICHENS
Frequency of lichens growing on
ground surfaces, logs, rocks, etc.
(# of 100-m2 plots in which
LICHENS occur/5 plots) x 100
C
freclarboreal
Frequency of arboreal Bryophytes
and Lichens
(# of 100-m2 plots in which
ARBOREA occur/5 plots) x 100
C
freclalgae
Frequency of filamentous or mat
forming algae
(# of 100-m2 plots in which ALGAE
occurs/5 plots) x 100
C
FRECL
MACROALGAE
Macroalgae (freshwater
species/seaweeds)
(# of 100-m2 plots in which
MACROALGA occurs/5 plots) x
100
C
XCOV_
BRYOPHYTES
Mean absolute cover bryophytes
growing on ground surfaces, logs,
rocks, etc.
S cover of 1RYOPHYTE across 5
plots/5 plots
C
XCOV_LICHENS
Mean absolute cover lichens
growing on ground surfaces, logs,
rocks, etc.
S cover of ICHENS across 5
plots/5 plots
C
XCOV_ARBOREAL
Mean absolute cover arboreal
Bryophytes and Lichens
2 cover of lRBOREAI across 5
plots/5 plots
C
XCOV_ALGAE
Mean absolute cover filamentous or
mat forming algae
2 cover of lLGAE across 5 plots/5
plots
C
XCOV_
MACROALGAE
Mean absolute cover macroalgae
(freshwater species/seaweeds)
2 cover of /IACROALGAE across 5
plots/5 plots
C
IMP_
BRYOPHYTES
Bryophytes growing on ground
surfaces, logs, rocks, etc.
(FRECLBRYOPHYTES +
XCOV_BRYOPHYTES)/2
C
IMP_LICHENS
Lichens growing on ground
surfaces, logs, rocks, etc.
(FREQJ.ICHENS +
XCOV_LICHENS)/2
C
IMP_ARBOREAL
Arboreal Bryophytes and Lichens
(FRECLARBOREAL +
XCOV_ARBOREAL)/2
C
IMP_ALGAE
Filamentous or mat forming algae
(FRECLALGAE + XCOV_ALGAE)/2
C
IMP_
MACROALGAE
Macroalgae (freshwater
species/seaweeds)
(freclmacroalgae +
XCOV_MACROALGAE)/2
C
Section 8
Ground Surface Attributes
Section 8.1
Water Cover and Depth
XH20_DEPTH
Mean Predominant water depth in
plots where water occurs
I 'REDOMINANT_DEPTH across
plots where standing water
occurs/number of plots where
standing water occurs
C
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
XH20_DEPTH_AA
Mean Predominant water depth
PREDOMINANT_DEPTh across
across AA
plots all sampled 100-m2 plots/5
plots
C
FREQJH20
Frequency of occurrence of water
(# of 100-m2 plots in which
across 100-m2 plots
TOTAL_WATER occurs/5 plots) x
100
C
MIN_C0V_H20
Minimum cover of water
Lowest value for TOTAL_WATER
across five 100-m2 plots
C
MAX_C0V_H20
Maximum cover of water
Highest value for OTAL_WATER
across five 100-m2 plots
C
XC0V_H20
Mean total cover of water (mean
S cover of TOTAL WATER acr< ss
percent of Veg Plot area with
5 plots/5 plots
C
water)
IMP_H20
Importance total cover of water
across Veg Plot area
(FRECLH20 + XC0V_H20)/2
C
Section 8.2
Bare ground and Vegetation Litter
UTTER_TYPE
Predominant litter type
PREDOMINANTJJTTER:
CONIFEROUS, DECIDUOUS,
GRAMINOID, FORB, FERN,
BROADLEAF
C
XDEPTH UTTER
Mean depth of litter across all 1-m2
Sum DEPTH SW and DEPTH NE
quadrats in AA
for all 1-m2 quadrats/total number
of sampled quadrats in AA (usually
10)
C
MEDDEPTH
Median depth of litter across all 1-
Median >EPTH SW and
LITTER
m2 quadrats in AA
DEPTH_NE for all 1-m2
quadrats/total number of sampled
quadrats in AA (usually 10)
C
FREQ_LITTER
Frequency of litter
(# of 100-m2 plots in which
TOTAL_LITTER >0/5 plots) x 100
C
freclbaregd
Frequency of bare ground
(# of 100-m2 plots in which any
one of EXPOSED_SOIL;
EXPOSED_GRAVEL;
EXPOSED_ROCK occurs/5 plots) x
100
C
freclexposed_
Frequency exposed soil/sediment
(# of 100-m2 plots in which
SOIL
EXPOSED_SOIL occurs/5 plots) x
100
C
FRECLEXPOSED_
Frequency exposed gravel/cobble
(# of 100-m2 plots in which
GRAVEL
(~2mm to 25cm)
EXPOSED_GRAVEL occurs/5 plots)
x 100
C
freclexposed_
Frequency exposed rock (> 25cm)
(# of 100-m2 plots in which
ROCK
EXPOSED_ROCK occurs/5 plots) x
100
C
FRECLWD_FINE
Frequency of fine woody debris (<
(# of 100-m2 plots in which
r
5cm diameter)
WD_FINE occurs/5 plots) x 100
159
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
FRECLWD_
Frequency of coarse woody debris
(# of 100-m2 plots in which
r
COARSE
(> 5cm diameter)
WD_COARSE occurs/5 plots) x 100
XCOV_UTTER
Mean Cover of litter
S cover of "OTAL_LITTER aci ss 5
plots/5 plots
c
XCOV_BAREGD
Mean cover of bare ground
S cover of :XPOSED_SOIL+
EXPOSED_GRAVEL +
EXPOSED_ROCI across 5 plots/5
plots
c
XCOV_EXPOSED_
Mean Cover exposed soil/sediment
S cover of iXPOSED SOIL acre ss 5
r
SOIL
plots/5 plots
XCOV_EXPOSED_
Mean Cover exposed gravel/cobble
S cover of iXPOSED GRAVEL
c
GRAVEL
(~2mm to 25cm)
across 5 plots/5 plots
XCOV_EXPOS E D_
c) Cover exposed rock (> 25cm)
S cover of iXPOSED ROCK acr ss
r
ROCK
5 plots/5 plots
XCOV_WD_FINE
Mean Cover of fine woody debris (<
S cover of WD FIN across 5
c,
5cm diameter)
plots/5 plots
in EW-VMMI
XCOV WD
Mean Cover of coarse woody debris
S cover of WD COARS across 5
r
COARSE
(> 5cm diameter)
plots/5 plots
L
IMPJJTTER
Importance of litter
(FREQJJTTER + XCOV_LITTER)/2
c
IMP_BAREGD
Importance of bare ground
(FRECLBAREGD +
XCOV_BAREGD)/2
c
IMP_EXPOSED_
Importance exposed soil/sediment
(freclexposed_soil +
r
SOIL
XCOV_EXPOSED_SOIL)/2
IMP_EXPOSED_
Importance exposed gravel/cobble
(frclexposed_gravel +
r
GRAVEL
(~2mm to 25cm)
XCOV_EXPOS E D_G RAVE L)/2
IMP_EXPOSED_
Importance exposed rock (> 25cm)
(freclexposed_rock +
r
ROCK
XCOV_EXPOS E D_ROCK)/2
IMP_WD_FINE
Importance of fine woody debris (<
(FRECLWD_FINE +
r
5cm diameter)
XCOV_WD_FINE)/2
IMP_WD_
Importance of coarse woody debris
(FRECLWD_COARSE+
r
COARSE
(> 5cm diameter)
XCOV_WD_COARSE)/2
SECTIONS 9- 11
METRICS BASED ON RAW DATA FROM FORM V-4: NWCA 2016
SNAG AND TREE COUNTS AND TREE COVER
Snag and tree metrics are calculated as means/100-m2 plots to represent
AA, unless specified as totals across AA (from all 5 100m2). Snag and tree
metrics were not placed on a per hectare basis because the AA and
sampled plots do not necessarily represent homogenous patches and
many wetlands are not forested but may have occasional trees. Basal
area was not calculated because diameters were estimated in classes.
SECTION 9
DEAD/SNAG COUNT METRICS -
Based on data from FORM V-4
(Snag/standing dead tree section)
TOTN_XXTHIN_
Total Number Dead tree or snags 5
I number of
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
TOTN XTHIN
Total number of dead trees or snags
2 number of 2 to
Count unique tree species (taxa) in
5m tall
LMED_TREE height class across all
5 plots
C
N_HMED_TREE
Richness tree species, trees > 5m to
Count unique tree species (taxa) in
15m tall
HMED_TREE height class across all
5 plots
C
N_TALL_TREE
Richness tree species, trees > 15m
Count unique tree species (taxa) in
to 30m tall
TALL_TREE height class across all 5
plots
C
161
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
N_VTALL_TREE
Richness tree species, trees > 30m
Count unique tree species (taxa) in
tall
VT_TREE height class across all 5
plots
C
N_TREE_
Richness tree species in ground
Count unique tree species (taxa) in
GROUND
layer (e.g., seedlings, saplings),
GROUND LAYER ( SMALL.TREE
r
trees < 2m
and SMALL_TREE height classes)
across all 5 plots
N_TREE_MID
Richness tree species in subcanopy
Count unique tree species (taxa) in
layer, trees 2m to 15m tall
MID LAYER (.MED_TREE and
HMED_TREE hei| ht classes) across
all 5 plots
c
N_TREE_UPPER
Richness tree species in subcanopy
Count unique tree species (taxa) in
layer, trees > 15m
UPPER LAYER (TALL_TREE and
VTALL_TREE height classes) across
all 5 plots
c
PCTN_TREE_
Percent richness of tree species
(N_TREE_GROUND/N_TREESPP) x
GROUND
found in ground layer (e.g.,
seedlings, saplings), trees < 2m
100
c
PCTN_TREE_MID
Percent richness of tree species
found in subcanopy layer, trees 2m
to 15m tall
(N_TREE_MID/N_TREESPP) x 100
c
PCTN_TREE_
Percent richness of tree species
(N_TREE_UPPER/N_TREESPP) x
UPPER
found in subcanopy layer, trees >
15m
100
c
FRECLVSMALL_
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
TREE
VSMALL trees, trees < 0.5m tall
any species of k/SMALL trees
occurs/5 plots) x 100
c
FREQ_SMALL_
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
TREE
SMALL trees, trees 0.5m to 2m tall
any species of SMALL trees
occurs/5 plots) x 100
c
FREQ_LMED_
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
TREE
LMED trees, trees > 2 to 5m tall
any species of LMED trees
occurs/5 plots) x 100
c
freclhmed_
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
TREE
HMED, trees > 5m to 15m tall
any species of HMED trees
occurs/5 plots) x 100
c
FREQ_TALL_TREE
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
TALL trees, trees > 15m to 30m tall
any species of TALL tree: occurs/5
plots) x 100
c
FREQ_VTALL_
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
TREE
Frequency of individual, trees >
30m tall
any species of i/TALL trees
occurs/5 plots) x 100
c
FREQ_TREE_
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
GROUND
ground layer trees < 2m
any species of GROUND LAYER
( fSMALL or SMALL) trees
occurs/5 plots) x 100
c
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
FREQ_TREE_MID
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
subcanopy, trees 2m to 15m tall
any species of MID LAYER ( U1ED
or HMED) trees occurs/5 plots) x
100
C
FREQ_TREE_
Frequency (proportion of plots) of
(Number of 100-m2 plots in which
UPPER
CANOPY trees, trees >15m
any species of UPPER LAYER
(LMED or HMED) trees occurs/5
plots) x 100
C
XCOV VSMALL
Mean absolute cover VSMALL trees,
2 of cover for all tree species in
TREE
trees < 0.5m tall
VSMALI height class across all
plots/5 plots
C
XCOV SMALL
Mean absolute cover SMALL trees,
2 of cover for all tree species in
TREE
trees 0.5m to 2m tall
SMAL height class across all
plots/5 plots
C
XCOV LMED
Mean absolute cover LMED trees,
2 of cover for all tree species in
TREE
trees > 2 to 5m tall
LMED height class across all
plots/5 plots
C
XCOV HMED
Mean absolute cover HMED trees,
2 of cover for all tree species in
TREE_
trees > 5m to 15m tall
HMEC height class across all
plots/5 plots
C
XCOV TALL TREE
Mean absolute cover TALL trees,
2 of cover for all tree species in
trees > 15m to 30m tall
TALI height class across all plots/5
plots
C
XCOV VTALL
Mean absolute cover VTALL trees,
2 of cover for all tree species in
TREE_
trees > 30m tall
VTALL height class across all
plots/5 plots
C
XCOV TREE
Mean absolute cover trees in
2 of cover for all tree species in
GROUND
ground layer (e.g., seedlings,
GROUND LAYER (VSMALL.TREE
r
saplings), trees < 2m
and SMALL_TREE height classes)
across all plots/5 plots
XCOV TREE MID
Mean absolute cover trees in MID
2 of cover for all tree species in
layer, trees 2m to 15m tall
MID LAYER (LMED_TREE and
HMED_TRE height classes) across
all plots/5 plots
c
XCOV TREE
Mean absolute cover trees in
2 of cover for all tree species in
UPPER
UPPER layer, trees >15m
UPPER LAYER (TALL_TREE and
VTALL_TREE height classes) across
all plots/5 plots
c
IMP_VSMALL_
Importance of VSMALL trees, trees
(FRECLVSMALL_TREE +
r
TREE
< 0.5m tall
XCOV_VSMALL_TREE)/2
IMP_SMALL_TREE
Importance of SMALL trees, trees
(FRECLSMALL_TREE +
r
0.5m to 2m tall
XCOV_SMALL_TREE)/2
IMP_LMED_TREE
Importance of LMED trees,trees > 2
(FREQJ_MED_TREE +
r
to 5m tall
XCOV_LMED_TREE)/2
L,
IMP_HMED_TREE
Importance of HMED trees, trees >
(freclhmed_tree +
r
5m to 15m tall
XCOV_HMED_TREE)/2
L,
IMP_TALL_TREE
Importance of TALL trees, trees >
(FREQJ"ALL_TREE +
r
15m to 30m tall
XCOV_TALL_TREE)/2
L,
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
IMP_VTALL_TREE
Importance of VTALL trees, trees >
(FRECLVTALL_TREE +
c
30m tall
XCOV_VTALL_TREE)/2
IMP_TREE_GROUN
Importance of trees in GROUND
(FREQJ"REE_GOUND +
D
layer (e.g., seedlings, saplings),
trees < 2m
XCOV_TRE E_G ROU N D)/2
c
IMP_TREE_MID
Importance of trees in MID layer,
(FREQJ"REE_MID +
r
trees 2m-15m tall
XCOV_TREE_MID)/2
IMP_TREE_UPPER
Importance of trees in UPPER layer,
(FREQJ"REE_UPPER +
r
trees > 15m
XCOV_TREE_UPPER)/2
SECTION 10.2
TREE COUNT METRICS
TOTN XXTHIN
Total number of tree stems in
2 number of tree stems in
TREE
XXTHIN class, trees 5 to 10cm DBH
(diameter breast height)
XXTHIN_TREE cla s across all
species and across all 100-m2 plots
c
TOTN XTHIN
Total number of tree stems in
2 number of tree stems in
TREE
XTHIN class, trees 11 to 25cm DBH
XTHIN_TREE class across all
species and across 100-m2 plots
c
TOTN THIN
Total number of tree stems in THIN
2 number of tree stems in
TREE
class, trees 26 to 50cm DBH
THIN_TREE class across all species
and across all 100-m2 plots
c
TOTN JR TREE
Total number of tree stems in JR
2 number of tree stems in
class, of trees 51 to 75cm DBH
JR_TREE cla s across all species
and across all 100-m2 plots
c
TOTN THICK
Total number of tree stems in THICK
2 number of tree stems in
TREE
class, trees 76 to 100cm DBH
THICK_TREE cla s across all
species and across all 100-m2 plots
c
TOTN XTHICK
Total number of tree stems in
2 number of tree stems in
TREE
XTHICK class, trees 101 to 200cm
DBH
XTHICK_TREE class across all
species and across all 100-m2 plots
c
TOTN XXTHICK
Total number of tree stems in
2 number of tree stems in
TREE
XXTHICK class, of trees > 200cm
DBH
XXTHICK_TREE class across all
species and across all 100-m2 plots
c
TOTN TREES
Total number of tree stems across
2 number of tree stems across all
all classes DBH
size classes, across all species, and
across all 100-m2 plots
c
TOTN LARGE
Total number of tree stems > 76cm
TOTN THICK TREE +
DBH
TOTN_XTHICK_TREE +
TOTN XXTHICK TREE
c
TOTN MID
Total number of tree stems 26 to
TOTN THIN TREE +
r
75cm DBH
TOTN_JR_TREE
L
TOTN SMALL
Total number of tree stems 5 to
TOTN XX THIN TREE +
r
25cm DBH
TOTN_XTHIN_TREE
L
XN_XXTHIN_
Mean number of tree stems in
TOTN_XXTHIN_TREES/5 plots
TREE
XXTHIN class, trees 5 to 10 cm DBH
(diameter breast height)
c
XN_XTHIN_TREE
Mean number of tree stems in
XTHIN class, trees 11 to 25cm DBH
TOTN_XTHIN_TREES/5 plots
c
164
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
CALCULATION (listed in Metric
Row),
SPECIES TRAIT TYPE (indicated in
Banner if applicable)
METRIC TYPE
METRIC NAME
METRIC DESCRIPTION
(C = condition,
S = stress)
XN_THIN_TREE
Mean number of tree stems in THIN
class, trees 26 to 50cm DBH
TOTN_TH 1 N_TREES/5 plots
C
XN_JR_TREE
Mean number of tree stems in JR
class, of trees 51 to 75cm DBH
TOTN_JR_TREES/5 plots
C
XN_THICK_TREE
Mean number of tree stems in
THICK class, trees 76 to 100cm DBH
TOTN_THICK_TREES/5 plots
C
XN XTHICK
Mean number of tree stems in
TOTN_XTHICK_TREES/5 plots
TREE
XTHICK class, trees 101 to 200 cm
DBH
C
XN XXTHICK
Mean number of tree stems in
TOTN_XXTHICK_TREES/5 plots
TREE
XXTHICK class, of trees > 200 cm
DBH
C
XN_TREES
Mean number of tree stems across
all classes DBH
TOTN_TREES/5 plots
C
XN LARGE
Mean number of tree stems > 76cm
XN THICK TREE +
DBH
XN_XTHICK_TREE +
XN_XXTHICK_TREE
C
XN_MID
Mean number of tree stems 26 to
XN_THIN_TREE + XN_JR_TREE
c
75cm DBH
L
XN_SMALL
Mean number of tree stems 5 to
XN_XX_THIN_TREE +
c
25cm DBH
XN_XTHIN_TREE
L
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
165
-------
Chapter 9: Vegetation Multimetric Indices and Wetland Condition
9.1 Overview - Vegetation
Multimetric Index (VMMI)
Multimetric indices (MMIs) of
ecological condition based on
biological assemblages (e.g.,
wetland vegetation, fish, birds,
periphyton, macroinvertebrates)
are cornerstones of the USEPA
National Aquatic Resource Surveys
(NARS). For MMIs, good and poor
condition are defined relative to
characteristics of the biota in least-
disturbed sites. This chapter
describes the process of
Vegetation Multimetric Index
(VMM!)development and the
development of thresholds (also
known as benchmarks) for good,
fair, and poor condition based on
VMMI values observed at least-
disturbed sites. Figure 1-1 in the
Analysis Overview illustrates how
the VMMI fits into the NWCA Analysis Pathway: 1) steps supporting VMMI development (see Chapter
6:,Chapter 7:, and Chapter 8: for details), 2) VMMI development and the determination of condition
thresholds based on VMMI values (this chapter), and 3) the use of VMMI values, condition thresholds,
and site weights in estimating wetland area in good, fair, or poor ecological condition (see Chapter 15:).
Previously, a national-scale VMMI, based on four broadly applicable metrics, was developed for the 2011
NWCA (USEPA 2016a, USEPA 2016b, Magee et al. 2019a). However, the availability of the added data
from the 2016 survey made it possible to develop more specific, finer-scale VMMIs. Using vegetation data
from the 1,985 unique NWCA sites sampled in 2011 or 2016 and methods detailed in Magee et al. (2019),
numerous candidate VMMIs were generated for the following site groups:
• National scale - all sampled wetlands (Table 8-1)
• Five subpopulations based on RPT_UNIT_6 groups (Figure 8-2, Table 8-3): tidaIly-infIueneed
Estuarine Wetlands in coastal areas (TDL) and Inland Wetlands in Five NWCA Aggregated
Ecoregions (ICP, EMU, PLNS-ARW, and WVM)
• Four broad Wetland Group subpopulations (WETCLS_GRP (Table 8-4))
Characteristics of these groups are discussed in Section 8.3. Candidate vegetation metrics that passed
several screening tests (Section 8.5) for each site group were used in VMMI development. The methods
for VMMI development (Section 9.3) detailed in this chapter were applied to all the above VMMI site
166
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
groups or subpopulations. Metrics that passed screening for the national-scale, for the five
subpopulations based on RPT_UNIT_6, or for the four of WETCLS_GRP subpopulations, were scored
(standardized) (Section 9.3.1) and used in developing numerous candidate VMMIs by subpopulation
(Section 9.3.2). Evaluation of performance criteria (Section 9.3.2) for the candidate VMMIs for each of
these groups, indicated the WETCLS_GRP subpopulations had the strongest performance. Based on these
results, four final VMMIs were ultimately selected, one for each Wetland Group: Estuarine Herbaceous
(VMMI-EH), Estuarine Woody (VMMI-EW), Inland Herbaceous (VMMI-PRLH), and Inland Woody (VMM-
PRLW). Thresholds for good, fair, and poor condition (Section 9.3.3) were established for each Wetland
Group VMMI. The Wetland Group VMMIs and their condition thresholds were used to calculate
population estimates of condition for the 2016 survey and for change analysis between 2011 and 2016
(USEPA 2022). These final four VMMIs are discussed in the results sections of this chapter (Sections 9.4
and 9.4.3). The NWCA 2021 survey used the same four VMMIs and condition thresholds developed for
the NWCA 2016 to calculate population estimates of conditon and for change analysis between 2021 and
earlier survey years.
The R-code for VMMI development and threshold assignment was developed using Statistical Software,
ver. 3.6.1 (R Core Team 2019).
9.2 Calibration and Validation Data
During the NWCA VMMI development process, numerous candidate vegetation metrics (n = 426, Section
8.8, Appendix E) were examined for potential utility in indicating condition and hundreds of thousands
potential VMMIs were generated and evaluated (Section 9.3). To aid in developing the strongest final
VMMIs and avoid over-fitting them to specific data collected in 2011 and 2016, vegetation data were
divided into calibration (80% of sampled sites, n = 1,587) and validation (20% of sampled sites, n = 398)
data sets. Numbers of calibration and validation sites in various subpopulations are listed in Table 8-1
through Table 8-4. Metric scoring and VMMI development were conducted using the calibration data and
the validation data were used to confirm the performance of the most promising candidate VMMIs.
The 20% of sampled sites included in the validation data were randomly selected from the total number
of sampled sites and reserved to evaluate the consistency of candidate VMMIs. To encompass the range
of disturbance and wetland types in the NWCA, sites for the validation data set were designated by
stratified-random selection based on disturbance class (least-, intermediate-, and most-disturbed) and
four Wetland Groups (WETCLS_GRP).
The 80% of sampled sites comprising the calibration data were used to score candidate metrics on a 0 to
10 continuous scale (Sections 9.3.1, 9.4). Candidate metrics that passed screening tests (Section 8.5) were
scored within the NWCA subpopulations used for metric screening and development of potential VMMIs.
The resulting metric scoring was applied to the corresponding validation data. A robust potential VMMI
based on calibration data metric scoring is expected to similarly distinguish least-disturbed from most-
disturbed sites for both calibration and validation data (VanSickle 2010, Magee et al. 2019a), and we
evaluated this ability using box-and-whisker plots (see Section 9.4)
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
167
-------
9.3 Developing Vegetation Multimetric Indices (VMMIs) - Methods
Using procedures that were developed for the 2011 NWCA (USEPA 2016b, Magee et al 2019), numerous
candidate VMMIs were generated and evaluated for the national scale, for subpopulations of
RPT_UNIT_6 (Table 8-3, Figure 8-2), and for subpopulations of WETCLS_GRP (Table 8-4). These groups
were selected in an effort to minimize within group variability and maintain a sufficient number of least-
disturbed sites within each group to allow VMMI development (see Section 8.3). Methods for candidate
VMMI generation and evaluation are summarized in this section and any differences from Magee et al.
2019 are incorporated in this summary.
9.3.1 Step 1 - Metric Scoring
Candidate metrics must be standardized to the same scale before they can be used as components of a
VMMI. Metrics that passed screening tests for a given subpopulation were standardized on a 0 to 10
continuous scale using the calibration data. The metrics were scored based on interpolation of metric
values between the 5th (floor) and 95th (ceiling) percentiles across all calibration sites (Blocksom 2003).
The direction of each metric was determined by the direction of the difference between the mean of the
least-disturbed sites and the mean of the most-disturbed sites. If the difference was positive, better
condition is associated with higher metric values, and if negative, the reverse is true. For metrics
decreasing with increasing disturbance, the ceiling was scored as 10 and the floor as zero. Conversely, for
metrics that increased with increasing disturbance, the floor was scored as 10 and the ceiling as zero.
Scores were truncated to 0 or 10 if observed values fell outside the floor to ceiling range. The resulting
metric scoring was applied to the corresponding validation (see Section 9.2) data. A robust potential
VMMI developed using this metric scoring should similarly distinguish least-disturbed from most-
disturbed sites for both the calibration and validation data.
9.3.2 Step 2 - Generating and Screening Candidate VMMIs
Determining the optimal set of metrics for inclusion in a Vegetation Multimetric Index (VMMI) is a
complex process. In analyses based on the 2011 NWCA vegetation data, USEPA (2016b) found that a
random approach for selecting sets of metrics to include in candidate VMMIs (adapted from VanSickle
2010) ultimately produced more robust VMMIs than did expert selection of sets of individual metrics that
were maximally responsive. Accordingly, Magee et al. (2019) refined this approach to build a national-
scale wetland VMMI using the 2011 NWCA vegetation data. The methods of Magee et al. (2019) were
applied to the vegetation data from 1,985 unique NWCA sites sampled in 2011 or 2016 to develop a set
of finer-scale VMMIs, based on NWCA subpopulations, for describing wetland condition across the
conterminous US. To this end, the calibration data set (n = 1,587 sites) was used to generate and evaluate
numerous candidate VMMIs.
Candidate VMMIs were developed based on all sites in the calibration data set for several NWCA
subpopulations using the final set of scored metrics applicable to that subpopulation. All candidate
metrics, passing metric screens for a particular subpopulation, were used in generating the random
metric combinations for the candidate VMMIs. Each potential VMMI was calculated and placed on a 100-
point scale using the formula:
VMMI = Z metric scores x 10/number of metrics
Magee et al. (2019) found that when developing VMMIs across numerous wetland types and large scales,
candidate VMMIs with between 4 and 6 metrics better distinguished least- and most-disturbed sites than
168
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
those with 7 to 10 metrics. Consequently, here, we considered candidate VMMIs based on 4, 5, or 6
randomly selected metrics for the national scale, the five RPT_UNIT_6 subpopulations and the four
WETCLS_GRP subpopulations. In addition, for the Inland Herbaceous (PRLH) and Inland Woody (PRLW)
Wetland Groups, candidate VMMIs based on 7 randomly selected metrics were also considered because
a somewhat larger number of metrics passed screening tests for these two groups. 100,000 candidate
VMMIs were generated for each metric number applicable to the national scale and each of the six
RPT_UNIT_6 subpopulations, and the four WETCLS_GRP subpopulations.
The resulting candidate VMMIs were evaluated using a series of performance criteria to determine which
VMMIs were most effective. Performance statistics for evaluating the candidate VMMIs included
measures of redundancy, sensitivity, repeatability, and precision (Magee et al. 2019). To avoid metric
redundancy, only candidate VMMIs with maximum and mean Pearson correlations among component
metrics of < 0.75 and < 0.5, respectively, were retained for further review. Sensitivity of each VMMI was
evaluated using an interval test, (Kilgour et al. 1998), alpha = 0.05, to determine the percentage of most-
disturbed sites with VMMI values that were significantly less than the fifth percentile of the distribution of
VMMI values for least-disturbed sites (Van Sickle 2010). Repeatability for each candidate VMMI was
assessed using a SignakNoise (S:N) ratio (Kaufmann et al. 1999, Section 8.5.2) calculated based on data
from the primary sampling visits for calibration sites and repeat sampling visits (i.e., revisits) for a subset
of these primary visit sites (see Table 8-1 through Table 8-4 for site numbers by subpopulations). The
standard deviation (SD) of VMMI values among the least-disturbed sites was used to describe precision of
the VMMI.
To identify the most effective candidate VMMIs for each subpopulation, we first arranged all relevant
VMMIs that passed the correlation filter in order of increasing correlation and decreasing sensitivity.
Typically, the VMMIs with the lowest correlations were also the most sensitive. Next, for the most
sensitive VMMIs in each subpopulation set (up to several hundred), those with the lowest mean and
maximum correlation among component metrics were identified. Among these, the VMMIs with the
highest S:N and smallest SD were identified. The resulting reduced set of candidate VMMIs was further
evaluated by collectively considering redundancy, sensitivity, S:N, SD, and the ecological content of
component metrics to identify 9 to 12 highest performing VMMIs in each VMMI subpopulation. Finally,
for each subpopulation, box-and-whisker plots were created for the set of 9 to 12 most promising VMMIs
to evaluate their robustness and responsiveness by comparing how well each VMMI distinguished 1) the
least- and most-disturbed sites for calibration data vs. validation data and 2) least- and most-disturbed
sites from data for all sampled sites in the pertinent subpopulation.
Consideration of the performance criteria and the box plots for the best candidate VMMIs informed the
selection of the four final VMMIs for use in condition assessment for NWCA 2016. The four VMMIs were
based on the Wetland Group subpopulations (Section 9.4): Estuarine Herbaceous (VMMI-EH), Estuarine
Woody (VMMI-EW), Inland Herbaceous (VMMI-PRLH), and Inland Woody (VMMI-PRW). Thresholds for
good, fair, and poor ecological condition for each of these VMMIs were set using the distribution of
VMMI values for subpopulation relevant least-disturbed sites. Lastly, for the four selected VMMIs, a final
evaluation of VMMI responsiveness was conducted using two sample t-tests (Welsh t-test to account for
unequal variances and sample size, Welsh 1947) to compare mean VMMI values between all sampled
least- and most-disturbed sites occurring within each Wetland Group.
9.3.3 Step 3 - Determining Ecological Condition Thresholds Based on VMMI Values
Thresholds for good, fair, and poor ecological condition were determined only for the final four Wetland
Group VMMIs: Estuarine Herbaceous (VMMI-EH), Estuarine Woody (VMMI-EW), Inland Herbaceous
169
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
(VMMI-PRLH), and Inland Woody (VMMI-PRLW). Prior to setting condition thresholds for each of these
VMMIs, the relevant set of least-disturbed sites were evaluated for outlier VMMI values, and values
below the 25th percentile - 1.5*IQR (interquartile range) for a VMMI group were excluded in setting
thresholds. Ecological condition categories (good, fair, and poor) were defined based on the distribution
of VMMI values observed in least-disturbed sites in a particular Wetland Group, following the percentile
approach described in Paulsen et al. (2008). Good condition was defined by VMMI values greater than or
equal to the 25th percentile, fair condition ranged from the 5th up to the 25th percentile, and poor
condition was delimited as less than the 5th percentile of the least-disturbed sites (Figure 9-1).
100
Least-Disturbed Site Distribution
Percentiles:
d)
L_
o
o
CO
(0
o
'¦p
CD
0
Q.
>»
1
Good Condition
Fair Condition
5 th
Poor Condition
Figure 9-1. Criteria for setting VMMI thresholds for good, fair, and poor condition categories based on VMMI
values observed for least-disturbed sites (REF_NWCA = L). Box plot whiskers: lower = the 25th percentile -1.5 X IQR
(interquartile range), upper = the 75th percentile + 1.5 X IQR.
Once the condition thresholds were established, each sampled site was assigned a condition category
(good, fair, or poor) based on the Wetland Group threshold applicable to the site and the site's observed
VMMI value.
9.4 Final 2016 VMMIs - Calculation, Performance and Condition Thresholds
Using the VMMI development process outlined in Section 9.3, four of the candidate VMMIs were selected
for use in estimating wetland area in good, fair, and poor condition based on the NWCA 2016 and for
estimating areal changes in wetland condition between NWCA 2011 and 2016. The four final VMMIs
represented subpopulations of WETCLS_GRP: Estuarine Herbaceous (EH), Estuarine Woody (EW), Inland
Herbaceous (PRLH), and Inland Woody (PRLW). These four VMMIs had stronger performance than the
highest-performing national-scale VMMI or the best VMMIs for subpopulations of RPT_UNIT_6.
Results describing the four final VMMIs for the NWCA 2016 analysis are organized under three headings:
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
• VMMI Description, Metric Scoring, and VMMI Calculation (Section 9.4.1)
• VMMI Performance (Section 9.4.2)
• VMMI Condition Thresholds (Section 9.4.3)
9.4.1 VMMI Description, Metric scoring, and VMMI Calculation
An overview of the Wetland Group VMMIs (VMMI-EH (Estuarine Herbaceous, VMMI-EW (Estuarine
Woody), VMMI-PRLH (Inland Herbaceous), and VMMI-PRLW (Inland Woody)) is provided in Table 9-1,
which lists the name and a brief description of all individual metrics included in each VMMI. Methods for
calculating the metrics comprising each VMMI can be found in Section 8.8:, Appendix E; metrics included
in the four NWCA 2016 Wetland Group VMMIs is indicated in the METRIC TYPE column of Appendix in
bold, color-coded font:
• VMMI-EH in light blue,
• VMMI-EW in dark blue,
• VMMI-PRLH in purple, and
• VMMI-PRLW in forest green.
Note that the Appendix E descriptions/formulas for how to calculate individual metrics may contain
names of other metrics listed in Appendix E or parameter names (Section 7.12:, Appendix C) that refer to
specific field collected data. For metrics that include information using species traits (e.g., growth habit,
duration, plant categories, wetland indicator status, native status, and coefficients of conservatism), it
may be useful to refer to the relevant section in Chapter 7: (Sections 7.6 through 7.9).
The NWCA metric scoring process (see Section 9.3), standardizes all individual metrics on a continuous
scale from 0 to 10, with higher values reflecting less disturbed conditions. Scoring of the metrics
comprising each VMMI was based on the metric values from the calibration data sites for that particular
VMMI site group (Table 8-4) and was applied to all sampled sites5 evaluated for that group. For scoring
the individual metrics that make up each VMMI (VMMI-EH, Table 9-2; VMMI-EW, Table 9-3; VMMI-PRLH,
Table 9-4; VMMI-PRLW, Table 9-5), the following information is provided 1) the direction of each metric's
response to disturbance based on observed metric values, 2) the metric floor (5th percentile) and ceiling
(95th percentile) values, and 3) the formula for metric scoring. VMMI-EH, VMMI-EW, and VMMI-PRLH
include one or metrics where observed values increase in response to disturbance. For metrics that
increase in response to disturbance, scoring is reversed so that the standardized metric scores will always
reflect less disturbance with higher values. The metric scoring reflected in Table 9-2 through Table 9-5
was used in to calculate VMMI values (scaled from 0 to 100) for each site based on the relevant VMMI for
the site (nwca_2016_veg_mmi.csv). The equations for VMMI-EH, VMMI-EW, VMMI-PRLH, and VMMI-
PRLW are presented immediately below the relevant scoring table.
5 All sampled sites include all Index Visits (probability and handpicked) and all site visits for sites sampled more than
once (i.e., revisit and resample events).
171
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 9-1. Metrics included in each of the four NWCA 2016 Vegetation Multimetric Indices (VMMIs). See Section
8.8:, Appendix E for formulas for calculation of these metrics.
VMMI
Metric Name
Description
EH
XRCOV_HTOL
Relative cover highly tolerant species (C-value <= 2)
(Estuarine Herbaceous) XRCOV_MONOCOTS_
.NAT
Relative cover native monocots
XRCOV SEN
Relative cover sensitive species (C-value >= 7)
XRCOV FORB
Relative cover forbs
N_ANNUAL
Annual species richness
PCTN NATSPP
Percent richness native species
EW
PCTN MONOCOT
Monocot percent richness
(Estuarine Woody)
XRCOV_GRAMINOID
Relative cover graminoids
RIMP_NATSPP
Relative importance native species
XCOV_WD_FINE
Mean Cover of fine woody debris (< 5cm diameter)
PCTN NATSPP
Percent richness native species
PCTNJSEN
Percent richness intermediately sensitive species (C-
value = 5 or 6)
PRLH
PCTN OBL FACW
Percent richness Obligate + Facultative Wetland
(Inland Herbaceous)
species
FQAI_ALL
Floristic quality index based on all species
XRCOV NATSPP
Relative cover native species
N TOL
Richness tolerant species (C-value < = 4)
PRLW
XRCOV MONOCOTS
NAT
Relative cover native monocots
(Inland Woody)
XC_ALL
Mean coefficient of conservatism based on all species
XRCOV NATSPP
Relative cover native species
RFREQJMATSPP
Relative frequency native species
Table 9-2. VMMI-EH metrics: floor and ceiling values, disturbance response, and interpolation formula for scorir
individual metrics. Final scores for each metric decrease with disturbance.
VMMI-EH Metrics
Unscored response Floor Ceiling Scoring formula (Observed = metric
to Disturbance
value at a given site)
XRCOV HTOL
Increases3
0
84.57 (84.57 - Observed)/(84.57 - 0)*10
XRCOV MONOCOTS
NAT Decreases
0.29 100 (Observed-0.29)/(100-0.29)*10
XRCOV_SEN
Decreases
0
100 (Observed-0)/(100-0)* 10
XRCOV FORB
Increases3
0
69.37 (69.37 - Observed)/(69.37 - 0)*10
N ANNUAL
Increases3
0
2 (2 - Observed)/(2 - 0)*10
PCTN_NATSPP
Decreases
62.96 100 (Observed-62.96)/(100-62.96)*10
Note: Scoring based on EH calibration data (n= 298 sites) and applied to all EH data (n = 374 sites).
aScoring is reversed for metrics that increase with disturbance. Scores truncated to 0 or 10 if observed values
fell outside the floor to ceiling range. Metrics are defined in Table 9-1.
The Estuarine Herbaceous VMMI (VMMI-EH) was calculated for each site on a continuous 0 to 100 scale:
VMM I EH = (XRC0V_HT0L_SC + XRC0V_M0N0C0TS_NAT_SC + XRCOV_SEN_SC
10
+ XRC0V_F0RB_SC + N_ANNUAL_SC + PCTN_NATSPP_SC) * —
6
where, the '_SC' suffix is the scored value for a metric.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 9-3. VMMI-EW metrics: floor and ceiling values, disturbance response, and interpolation formula for scoring
individual metrics. Final scores for each metric decrease with disturbance.
VMMI-EW Metrics Unscored response Floor Ceiling Scoring formula (Observed = metric
to Disturbance value at a given site)
PCTN MONOCOT Decreases
0 55.68 (Observed-0)/(55.68-0)* 10
XRCOV_GRAMINOID Decreases 0 90.36 (Observed-0)/(90.36-0)* 10
RIMP NATSPP Decreases
68.56 100 (Observed-68.56)7(100-68.56)*10
XCOV_WD_FINE Increases3 0 13.85 (13.85 - Observed)/(13.85-0)*10
PCTN NATSPP Decreases
66.98 100 (Observed-66.98)7(100-66.98)*10
PCTNJSEN Decreases 7.57 45.45 (Observed - 7.57)7(45.45 - 7.57)* 10
Note: Scoring based on EW calibration data (n = 70 sites) and applied to all EW data (n = 87 sites).
aScoring is reversed for metrics that increase with disturbance. Scores truncated to 0 or 10 if observed values
fell outside the floor to ceiling range. Metrics are defined in Table 9-1.
The Estuarine Woody VMMI (VMMI-EW) was calculated for each site on a continuous 0 to 100 scale:
VMMI EW = (PCTN_M0N0C0T _SC + XRCOV_GRAMINOID_SC + RIMP_NATSPP_SC
10
+ XCOV_WD_FINE_SC + PCTN_NATSPP_SC + PCTN_ISEN_SC) * —
6
where, the '_SC' suffix is the scored value for a metric.
Table 9-4. VMMI-PRLH metrics: floor and ceiling values, disturbance response, and interpolation formula for
scoring individual metrics. Final scores for each metric decrease with disturbance.
VMMI-PRLH Metrics Unscored response Floor Ceiling Scoring formula (Observed = metric value
to Disturbance at a given site)
PCTN_OBL_FACW Decreases 17.21 100 (Observed -17.21)7(100 -17.21)*10
FQAI_ALL Decreases 4.90 35.77 (Observed-4.90)7(35.77-4.90)* 10
XRCOV NATSPP Decreases
12.42 100 (Observed - 12.42)7(100 - 12.42)*10
N_TOL Increases3 3 41 (41 - Observed)/(41 - 3)*10
Note: Scoring based on PRLH calibration data (n = 522 sites) and applied to all PRLH data (n = 654 sites).
aScoring is reversed for metrics that increase with disturbance. Scores truncated to 0 or 10 if observed values
fell outside the floor to ceiling range. Metrics are defined in Table 9-1.
The Inland Herbaceous VMMI (VMMI-PRLH) was calculated for each site on a continuous 0 to 100 scale:
VMMI PRLH = (PCTN_OBL_FACW _SC + FQAI_ALL_SC + XRCOV_NATSPP_SC
N 10
+ N TOL SC) * —
4
where, the '_SC' suffix is the scored value for a metric.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
173
-------
Table 9-5. VMMI-PRLW metrics: floor and ceiling values, disturbance response, and interpolation formula for
scoring individual metrics. Final scores for each metric decrease with disturbance.
VMMI-PRLW Metrics Unscored response Floor Ceiling Scoring formula (Observed = metric
to Disturbance value at a given site)
XRCOV MONOCOTS NAT
Decreases
0.17
48.41
(Observed - 0.17)/(48.41 - 0.17)* 10
XC ALL
Decreases
2.52
6.19
(Observed - 2.52)/(6.19 - 2.52)*10
XRCOV_NATSPP
Decreases
53.04
100
(Observed - 53.04)/(100 - 53.04)* 10
RFRECLIMATSPP Decreases 62.83 100 (Observed - 62.83)/(100 - 62.83)* 10
Note: Scoring based on PRLW calibration data (n = 697 sites) and applied to all PRLW data (n = 872 sites).
aScoring is reversed for metrics that increase with disturbance. Scores truncated to 0 or 10 if observed
values fell outside the floor to ceiling range. Metrics are defined in Table 9-1.
The Inland woody VMMI (VMMI-PRLW) was calculated for each site on a continuous 0 to 100 scale:
VMM I PRLW = (XRC0V_M0N0C0TS_NAT_SC + XC_ALL_SC + XRCOV_NATSPP_SC
^ 10
+ RFREQ_NATSPP_SC) * —
where, the '_SC suffix is the scored value for a metric.
9.4.2 VMMI Performance
Descriptive statistics - Descriptive statistics for the Wetland Group VMMIs (EH, EW, PRLH, PRLW) are
summarized in Table 9-6. The high S:N values for the EH, PRLH, and PRLW VMMIs reflect consistency in
the VMMI across repeat samplings. However, the S:N value for the EW VMMI is not very meaningful
because only two revisit sites (i.e., a second sampling visit to a site during the same year as the first
sampling visit) were available. Low mean correlations among metrics in VMMI indicate low redundancy
among metrics. Sensitivity, or the percentage of most-disturbed sites distinguished from least-disturbed
sites, based on the conservative Kilgourtest (VanSickle 2010), varies by wetland type group. The observed
sensitivity values were comparatively high for MMIs (see Magee et al. 2019). VMMI-PRLW had the lowest
separation of least- and most-disturbed sites, a pattern that may be influenced by the diversity of specific
wetland community types and structural types within the PRLW group.
Box plot comparisons of calibration and validation data by VMMI - For all four VMMIs, comparison of
VMMI values between calibration and validation data showed similar distributions and satisfactory
discrimination between least- and most-disturbed sites (top graph in Figure 9-2 (VMMI-EH), Figure 9-3
(VMMI-EW), Figure 9-4 (VMMI-PRLH), and Figure 9-5 (VMMI-PRLW)). Similar results between calibration
and validation data sets indicate consistent behavior for the VMMIs across different data sets, suggesting
robustness of VMMI performance for wetland data collected in future years. Sample sizes in the
validation data were very small for the Estuarine Woody VMMI (EW), so in this case differences in VMMI
values for least-disturbed sites between calibration and validation data may not be meaningful.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Box plot comparisons of least- and most-disturbed sites by VMM I - For each of the Wetland Group VMMIs
(bottom graphic in Figure 9-2 (VMMI-EH), Figure 9-3 (VMMI-EW), Figure 9-4 (VMMI-PRLH), and Figure 9-5
(VMMI-PRLW)), box plots comparing the VMMI value distributions showed clear separation for the set of
least- and most-disturbed unique sites sampled in 2011 and 2016. These figures illustrate no overlap in
VMMI values between the 25th percentile of the least-disturbed sites and the 75th percentile of most-
disturbed sites for VMMI-EH, VMMI-EW and VMMI-PRLH. The separation between least- vs. most-
disturbed sites is somewhat less distinct for the Inland Woody VMMI (PRLW), with slight overlap of the
25th percentile of the least-disturbed sites and the 75th percentile of most-disturbed sites. However, this
response was improved by separating the PRLW sites into two groups, arid vs. mesic, for setting condition
thresholds (see Section 9.4.3 and Figure 9-6).
Table 9-6. Summary statistics for the final four VMMIs: EH - Estuarine Herbaceous, EW -Estuarine Woody, PRLH -
Inland Herbaceous, PRLW - Inland Woody. Statistics calculated based on VMMI values for sampled sites and revisit
sites from the calibration data set for the relevant VMMI group.
VMMI
n-sites by
Mean
SD1
S:N2
Max r
Mean r
Sensitivity
n = calibration
disturbance
VMMI
VMMI
n = revisit
among
among
(%)
data sites3
class
(L sites)
(L sites)
sites4
metrics
metrics
EH
n=298
L=107, 1=126,
M=65, ? = 0
92.37
10.37
35.53
n = 16
0.63
0.39
61.53
EW
n=70
L=12, 1=34,
M=22, ? =2
75.52
7.46
32.12
n = 2
0.73
0.14
72.73
PRLH
n=522
L=77, 1=293,
M=150,1=2
78.27
11.46
16.82
n = 29
0.39
0.17
50.67
PRLW
n=697
L=155, 1=395,
M=143, ?=4
68.95
12.48
17.30
n = 36
0.73
0.37
32.17
VMMIs defined in Section 9.4.1. L = least disturbed sites, I = intermediately disturbed sites, M=most
disturbed sites, ?=undetermined disturbance, 1SD =standard deviation, 2S:N = signal/noise (For each VMMI,
S:N is based on the 3sampled sites and the 4revisit sites from calibration data set), r = Pearson correlation.
Sensitivity = Percent M sites with VMMI values significantly less than the fifth percentile of the distribution of
VMMI values for L sites based on an interval test, alpha = 0.05 (Kilgour et al. 1998, Van Sickle 2010).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
175
-------
Estuarine Herbaceous
100
x
0
TJ
C
75
| 50
I »
03
CD
(U
>
Calibration
Validation
E±] Least Disturbed ^1 Most Disturbed
Figure 9-2. Comparison of VMMI Estuarine Herbaceous wetlands (VMMI-EH) for least-disturbed and most-
disturbed sites. Top graph: Compares VMMI values for least- and most-disturbed EH sites in the calibration and
validation data sets. Bottom graph; VMMI values for all least- and most-disturbed sampled EH sites. Box plots: box
is interquartile (IQR) range, line in box is the median, and whiskers represent most extreme point a distance of no
more than 1.5 x IQR from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
176
-------
Estuarine Woody
100
x
0)
~o
c
75
g 50
i»
CD
¦*—¦
01
CS)
OJ
>
Calibration
Validation
$ Least Disturbed Most Disturbed
Estuarine Woody
100
x
Least Disturbed
Most Disturbed
Figure 9-3, Comparison of VMMI Estuarine Woody wetlands (VMMI-EW) for least-disturbed and most-disturbed
sites. Top graph: Compares VMMI values for least- and most-disturbed EW sites in the calibration and validation
data sets. Bottom graph: VMMI values for all least- and most-disturbed sampled EW sites. Box plots: box is
interquartile (IQR) range, line in box is the median, and whiskers represent most extreme point a distance of no
more than 1.5 x IQR from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
177
-------
Inland Herbaceous
$ Least Disturbed ^1 Most Disturbed
Figure 9-4. Comparison of VMMI Inland herbaceous wetlands (VMMI-PRLH) for least-disturbed and most-disturbed
sites. Top graph: Compares VMMI values for least- and most-disturbed PRLH sites in the calibration and validation
data sets. Bottom graph: VMMI values for all least- and most-disturbed sampled PRLH sites. Box plots: box is
interquartile (IQR) range, line in box is the median, and whiskers represent most extreme point a distance of no
more than 1.5 x IQR from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
178
-------
$ Least Disturbed Most Disturbed
Least Disturbed Most Disturbed
Figure 9-5. Comparison of VMMI Inland woody wetlands (VMMI-PRLW) for least-disturbed and most-disturbed
sites. Top graph: Compares VMMI values for least- and most-disturbed PRLW sites in the calibration and validation
data sets. Bottom graph: VMMI values for all least- and most-disturbed sampled PRLW sites. Box plots: box is
interquartile (IQR) range, line in box is the median, and whiskers represent most extreme point a distance of no
more than 1.5 x IQR from the box. Values beyond this distance are outliers. Numbers below each box plot
represent number of the least-disturbed or most-disturbed sites sampled.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
179
-------
9.4.3 Condition Thresholds for the Wetland Group VMMIs
Wetland condition thresholds for each the four final VMMIs (Table 9-7) were based on the distribution of
VMMI scores in least-disturbed sites (see Section 9.3.3):
• Good = VMMI scores > 25th percentile of least-disturbed sites
• Fair = VMMI scores from the 5th up to the 25th percentile of least-disturbed sites
• Poor = VMMI scores < 5th percentile of least-disturbed sites
• Least-disturbed sites in a Wetland Group with VMMI values below the 25th percentile - 1.5*IQR
(interquartile range) were considered outliers and not used in setting condition thresholds
Note that the VMMI-PRLW values in least-disturbed PRLW sites varied widely by ecoregion (Figure 9-6,
top graph). As a result, two sets of thresholds were developed for the VMMI-PRLW, one set for sites in
more mesic regions (PRLWother) and one set for sites in more arid regions (PRLWplnarw) (Table 9-7, Figure
9-6, bottom graph). A final evaluation of responsiveness (two sample unequal variance t-tests) for each of
the four NWCA VMMIs and the two VMMI-PRLW threshold groups showed significantly different mean
VMMI values between all sampled least- and most-disturbed sites (Table 9-8).
Each sampled site was assigned a condition category (good, fair, or poor) based on the site's observed
VMMI value and the Wetland Group VMMI and condition thresholds applicable to the site
(nwca_2016_veg_mmi.csv).
Table 9-7. VMMI value thresholds indicating good, fair, and poor ecological condition based on least-disturbed
sites in each Wetland Group (WETCLS_GRP). Sites with VMMI values from the 5th up to the 25th percentile for least-
disturbed (REF_NWCA) sites are considered in fair condition.
NWCA VMMIs
(n = least-
disturbed sites)
Poor Condition
Good Condition
Description (Wetland Type and Site Groups)
(VMMI < 5th
Percentile Least-
(VMMI > 25th
Percentile Least-
Disturbed Sites)
Disturbed Sites)
VMMI-EH
(n =134)
Tidal - Estuarine Flerbaceous [ALL]
73.6
86.4
VMM-EW
(n = 15)
Tidal - Estuarine Woody [ALL]
64.6
69.8
VMM-PRLH
Inland (Palustrine, Riverine, or Lacustrine)
Flerbaceous [ALL]
63.8
74.2
VMMI-PRLW
Inland (Palustrine, Riverine, or Lacustrine) Woody
PRLWother
(n = 157)
Inland (Palustrine, Riverine, or Lacustrine) Woody
[EMU, ICP, WVM]
53.7
65.5
PRLWplnarw
(n = 37)
Inland (Palustrine, Riverine, or Lacustrine) Woody
[PLN, ARW]
43.7
49.9
Table 9-8. Two-sample unequal variances t-tests comparing VMMI value means for all sampled least- and most-
disturbed sites for each Wetland Group VMMI.
-------
Inland Woody
100-
x
0)
"O
c
75
15 50
E
c
o
25
eTB
65 71
\G?
72 46
34 23
3 13
0^
20 26
$ Least Disturbed ^ Most Disturbed
Inland Woody
-100
x
0)
TJ
C
75
a3 50
E
25
0)
CD
O)
>
$ Least Disturbed ^ Most Disturbed
Figure 9-6. Comparison of VMMI values for Inland Woody wetlands (VMMI-PRLW) for least-disturbed and most-
disturbed sites by ecoregions. Top graph: VMMI-PRLW values by Five NWCA Aggregated Ecoregions (ICP, EMU,
PLN, ARW, WVM, (NWCA_EC05) see map in Figure 6-2 for definitions) Bottom graph: VMMI-PRLW values for
more mesic (OTHER) vs. more arid (PLN_ARIDW) regional groups (OTHER = ICP, EMU, WVM; PLN_ARIDW = ARW &
WVM). Box plots: box is interquartile (IQR) range, line in box is the median, and whiskers represent most extreme
point a distance of no more than 1,5 x IQR from the box. Values beyond this distance are outliers. Numbers below
each box plot represent number of the least-disturbed or most-disturbed sites sampled.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
9.5 Literature Cited
Blocksom KA (2003) A performance comparison of metric scoring methods for a multimetric index for
Mid-Atlantic Highlands streams. Environmental Management 31: 670-682
Kaufmann PR, Levine P, Robison EG, Seeliger C, Peck DV (1999) Quantifying Physical Habitat in Wadable
Streams. EPA/620/R_99/003. US Environmental Protection Agency, Washington, DC
Kilgour BW, Sommers KM, Matthews DE (1998) Using the normal range as a criterion for ecological
significance in environmental monitioring and assessment. Ecoscience 5: 542-550
Kincaid TM, Olsen AR, Weber MH (2019). spsurvey: Spatial Survey Design and Analysis. R package version
4.1.0.
Magee TK, Blocksom KA, Fennessy MS (2019) A national-scale vegetation multimetric index (VMMI) as an
indicator of wetland condition across the conterminous United States. Environmental Monitoring and
Assessment 191 (SI): 322, doi: 10.1007/sl0661-019-7324-4.
https://link.springer.com/article/10.1007/sl0661-019-7324-4
Paulsen SG, Mayio A, Peck DV, Stoddard JL, Tarquinio E, Holdsworth SM, Sickle JV, Yuan LL, Hawkins CP,
Herlihy AT, Kaufmann PR, Barbour MT, Larsen DP, Olsen AR (2008) Conditions of stream ecosystems in
the US: an overview of the first national assessment. Journal of the North American Benthological Society
27(4), 812-821
R Core Team (2019) R: A language and environment for statistical computing. Version 3.6.1. R Foundation
for Statistical Computing, Vienna, Austria. (http://www.R-project.org/)
Van Sickle J (2010) Correlated metrics yield multimetric indices with inferior performance. Transactions of
the American Fisheries Society 139: 1802-1817
USEPA (2016a) National Wetland Condition Assessment 2011: A Collaborative Survey of the Nation's
Wetlands. EPA-843-R-15-005. US Environmental Protection Agency, Office of Water, Washington, DC
USEPA (2016b) National Wetland Condition Assessment: 2011 Technical Report. EPA-843-R-15-006. .US
Environmental Protection Agency, Washington, DC. https://www.epa.gov/national-aquatic-resource-
surveys/national-wetland-condition-assessment-2011-results
USEPA (2022) National Wetland Condition Assessment 2016: The Second Collaborative Survey of Wetlands
in the United States. EPA-841-R-23-001. US Environmental Protection Agency, Office of Water,
Washington, DC
Welch BL (1947). "The generalization of "Student's" problem when several different population variances
are involved". Biometrika. 34 (1-2): 28-35
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
182
-------
Chapter 10: Nonnative Plant Indicator (NNPI)
10.1 Background
Nonnative plant species are widely recognized as important biological indicators of ecological stress on
wetland condition (Mack and Kentula 2010, Magee et al. 2010). They can 1) reflect ecological condition of
the 'natural' vegetation, 2) be indicators of anthropogenic disturbance, or 3) behave as direct stressors to
vegetation and ecosystem properties (e.g., Kuebbing et al. 2015, Magee et al. 2008, 2010, 2019, Pysek et
al. 2020, Riccardi et al. 2020, Ruaro et al 2020, Simberloff 2011). Presence and abundance of nonnative
plants are often positively related to human mediated disturbance (Lozon and Maclsaac 1997, Mack et al.
2000, Magee 1999, Magee et al. 2008, Ringold et al. 2008). In addition, nonnative plants can act as direct
stressors to ecological condition by competing with or displacing native plant species or communities,
altering vegetation structure, or by altering ecosystem structure and processes (Vitousek et al. 1997,
Dukes and Mooney 2004). Numerous direct and indirect effects of nonindigenous plants on native
vegetation and other ecosystem components demonstrate their role as potential stressors and indicators
of lowered ecological condition.
For example, nonnative plant species have been linked to:
• increased risk of local extinction or population declines for many rare, native plant species
(Randall 1996, Lesica 1997, Seabloom et al. 2006),
• changes in species composition within and among plant community types, and to local and
regional floristic homogenization (McKinney 2004, Rooney et al. 2004, Magee et al. 2008),
• alteration of fire regimes (Dwire and Kauffman 2003, Brooks et al. 2004),
• alteration of geomorphic and hydrologic processes (Rowantree 1991, Sala et al. 1996), and
• alteration of carbon storage patterns (Farnsworth and Meyerson 2003, Bradley et al. 2006),
nutrient cycling, and composition of soil biota (Belnap and Phillips 2001, Ehrenfeld 2003).
Major ecological changes like these negatively influence the intactness or integrity of natural ecosystems
(Angermeier and Karr 1994, Dale and Beyeler 2001) and can lead to losses of ecosystem services (Dukes
and Mooney 1999, Dale et al. 2000, Hooper et al. 2005, Meyerson and Mooney 2007).
Recall from Section 7.8 (Species Traits - Native Status) and from Magee et al. 2019 that NWCA defines
nonnative plants to include both alien and cryptogenic taxa. Concepts describing native status categories
used by the NWCA, including alien and cryptogenic, are described in brief here and in Table 7-5. First,
Native plant taxa are defined as indigenous to specific states in the conterminous US. Introduced taxa are
indigenous outside of, and not native, in any of conterminous US. Adventive taxa are native to some parts
of the conterminous US but introduced to the location of occurrence. We use the term Alien to include
both introduced and adventive taxa. Cryptogenic species include taxa that have both introduced (often
aggressive) and native (generally less prevalent) genotypes, varieties, or subspecies. Because many
cryptogenic species are invasive or act as ecosystem engineers (Magee et al. 2019), we grouped them
with alien species and considered them nonnative for the purpose of indicating ecological stress.
The Nonnative Plant indicator (NNP/f was developed as a categorical descriptor of stress to ecological
condition for the NWCA 2011 (Magee et al 2019, USEPA 2016a) and has been used in subsequent NWCA
6 In the NWCA 2011 Technical Report (USEPA 2016a), the NNPI was referred to as the "Nonnative Plant Stressor
Indicator" (NPSI) - a name that is no longer used.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
analyses. Magee et a I. (2019) detailed the development of the Nonnative Plant Indicator (NNPI) and,
based on the 2011 NWCA, reported on relationships of the NNPI to disturbance and environmental
conditions and on the 2011 extent of wetland area in different NNPI condition categories.
In the following subsections, data collection, data preparation, description of the NNPI, and condition
category threshold definitions are described.
10.2 Data Collection
Nonnative plant data were collected as part of the standard Vegetation Protocol (USEPA 2021a). An
overview of vegetation field and laboratory methods is provided in Chapter 7, Section 7.3.
10.3 Data Preparation
Preparation and validation of raw data for nonnative plant species are described in Chapter 1, Sections
7.4 and 7.5. Definition of the native status categories used in the NWCA and the procedures for
determining state-level native status for the individual species observed in the 2011, 2016 and 2021
NWCAs are provided in Chapter 7, Section 7.8. Numerous metrics summarizing different attributes of
nonnative species (e.g., all alien and cryptogenic species, or subgroups of these species based on life
history traits) were calculated and are described in Chapter 8, Sections 8.4 and 8.8 (Appendix E).
10.4 Nonnative Plant Indicator Overview
The categorical NNPI was based on three straightforward continuous metrics (Table 10-1) that reflect
different potential impacts of nonnative plants, and which can be readily calculated from field
observations.
Table 10-1. Definition of metrics used in the NNPI.
Metric Name
Calculation3
Range
VC - Relative Cover of
(2 Absolute cover nonnative species,/£ Absolute
0 to 100%
Nonnative Species
cover all species,) x 100; where for each unique
species /': Absolute Cover = 0-100%
TOTN_AC - Richness of Nonnative
Number of unique nonnative species observed at a
Number of unique
Species
site
nonnative species
RFREQ^AC - Relative Frequency of £ Frequency nonnative species,/£ Frequency all 0 to 100%
Nonnative Species species,) x 100; where for each unique species /':
Frequency = 0-100%, calculated as the percent of
Veg Plots in which it occurred.
Calculation of metrics based on data collected in the five 100-m2 vegetation plots sampled at each site.
Additional information about these metrics can be found in Chapter 8, Section 8.8 (Appendix E) by
referencing the metric names indicated in red font in the list above and highlighted in red and bolded in
184
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
the appendix. The "_AC" suffix in the metric names refers to combined alien and cryptogenic species that
together are considered nonnative by the NWCA.
Each of the three metrics consider all nonnative species at a location but, taken together, integrate
different avenues of impact to ecological condition. Relative Nonnative Cover[0 to 100%) reflects
preemption of space and resources and is often associated with changes in plant community composition
(species identity, richness, and abundance) and vegetation structure (horizontal or vertical), or with
alteration of ecosystem processes (e.g., hydrology, nutrient cycling, fire regime). Greater Richness of
Nonnative Species (number of unique nonnative species) increases the risk that individual nonnative taxa
are or may become invasive or act as ecosystem engineers that negatively alter biotic or abiotic
properties. Increasing Relative Nonnative Frequency (0 to 100%) across a site reflects increasing numbers
of loci from which nonnatives could compete with native species, expand in cover, or spread to new
locations. Of the three metrics, relative nonnative cover is likely to represent the greatest potential
negative effect on ecological condition. The other two metrics provide additional pathways of impact that
may have synergistic relationships with relative nonnative cover, potentially increasing the amount
overall stress related to nonnative plants.
The three metrics of the NNPI are used together in a decision matrix to assign a condition category
reflecting potential stress from nonnative species to each site. Four condition categories (good, fair, poor,
or very poor) were defined7. Assignment of the condition category for each site is based exceedance
values for each of the three metrics; see the following section (Section 8.5) for details.
10.5 NNPI Condition Threshold Definition
NNPI condition thresholds were developed to:
• reflect wetland condition as an additional indicator to the VMMI (Chapter 9:) and
• indicate stressor condition related to nonnative plants.
The same thresholds were used for both of these purposes. Details of how the NNPI is used in final
reporting for wetland condition and stressor condition are discussed in Chapter 15:.
The three NNPI metrics (nonnative relative cover, nonnative richness, and nonnative relative frequency),
were used together in a decision matrix to assign each sampled site to a condition category (good, fair,
poor, or very poor) based on exceedance values for each of the metrics (see Table 10-2, below, and
Magee et al. 2019). The overall NNPI status for each site was determined by the lowest condition
category observed across the three NNPI metrics.
Exceedance values for the four condition categories for each metric were developed by Magee et al.
(2019) using best professional judgement, considering diverse wetland community types and changes in
plant community composition and structure with varying levels of nonnative cover, frequency, or
richness. Exceedance values for the four condition categories (Table 10-2) reflect the strong influence of
7 In previous work (USEPA 2016a, Magee et al. 2019), the NNPI categories were described in relation to potential
stress (i.e., low, moderate, high, or very high). However, to better align with other USEPA National Aquatic Resource
Surveys, the NNPI categories were renamed to reflect condition (good, fair, poor, or very poor). Now, good
condition is equivalent to the previously defined low stressor-level, and very poor condition is equivalent to the
formerly described very high stressor-level.
185
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
nonnative relative cover, with the values for nonnative richness and nonnative relative frequency set to
consider these two metrics as additional sources of ecological stress.
Table 10-2. Condition Threshold Exceedance Values for each of the metrics informing the Nonnative Plant Indicator
(NNPI): Relative Cover of Nonnative Species (XRCOV_AC), Nonnative Richness (TOTN_AC), and Relative Frequency
of Nonnative Species (RFREQ_AC).
Condition Category*
XRCOV_AC
TOTN_AC
RFRECLAC
Good
<1
<5
<10
Fair
>1-15
>5-10
>10-30
Poor
>15-40
>10-15
>30-60
Very Poor
>40
>15
>60
* Exceedance of a threshold value for a particular condition category for any one of the three metrics moves the
metric condition to next lower (better to worse) category. The NNPI condition for a site is based on the lowest
observed condition category among the metrics.
The approach for designating the NNPI condition category for each site integrates information from three
different pathways from which nonnative species may influence ecological condition. To see how the
exceedance thresholds work, consider the two hypothetical examples of nonnative species results that
are outlined below.
Hypothetical Site 1 (NNPI Condition Category = Poor) has:
• XRCOV_AC = 7% ^ Fair Condition
• TOTN_AC = 14 nonnative species ^ Poor Condition
• RFREQ_AC = 52% ^ Poor Condition
Hypothetical site 1 has nonnative relative cover of 7%, placing the site in the fair condition category.
However, this site also has nonnative richness of 14 species and relative frequency of 52%, which reflect
poor condition for both metrics. Thus, the site would be assigned to the NNPI poor condition category.
Even though relative nonnative cover is not extensive at this hypothetical site, the number of individual
nonnative species and their frequency of occurrence might indicate shifting community composition and
strong risk for expansion of nonnative cover.
Hypothetical Site 2 (NNPI Condition Category = Very Poor) has:
• XRCOV_AC = 80% ^ Very Poor Condition
• TOTN_AC = 1 nonnative species ^ Good Condition
• RFREQ_AC = 59% ^ Poor Condition
Next, consider hypothetical site 2 with 80% nonnative relative cover indicating very poor condition,
nonnative richness of 1 indicating good condition, and nonnative relative frequency of 59% indicating
poor condition. Here, the overall NNPI condition category would be very poor. Even though there is only
one nonnative species present at the site, it occupies 80% of the total vegetation cover and nearly 60% of
all species occurrences across the sampled area of the vegetation plots are nonnative.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
186
-------
10.6 Literature Cited
Angermeier PL, Karr JR (1994) Biological integrity versus biological diversity as policy directives. Bioscience
44: 690-697
Belnap J, Phillips SL (2001) Soil biota in an ungrazed grassland: Response to annual grass (Bromus
tectorum) invasion. Ecological Applications 11: 1261-1275
Bradley BA, Houghton RA, Mustard JF, Hamburg SP (2006) Invasive grass reduces aboveground carbon
stocks in shrublands of the Western US. Global Change Biology 12: 1815-1822
Brooks ML, D'Antonio CM, Richardson DM, Grace JB, Keeley JE, DiTomaso JM, Hobbs RJ, Pel la nt M, Pyke D
(2004) Effects of Invasive Alien Plants on Fire Regimes. Bioscience 54: 677-688
Dale VH, Beyeler SC (2001) Challenges in the development and use of ecological indicators. Ecological
Indicators 1: 3-10
Dale VH, Brown SC, Haeuber RA, Hobbs NT, Huntly N, Naiman RJ, Riebsame WE, Turner MG, Valone TJ
(2000) Ecological principles and guidelines for managing the use of land. Ecological Applications 10: 639-
670
Dukes JS, Mooney HA (1999) Does global change increase the success of biological invaders? Trends in
Ecology and Evolution 14: 135-139
Dukes JS, Mooney HA (2004) Disruption of ecosystem processes in western North America by invasive
species. Revista Chilena de Historia Natural 77: 411-437
Dwire KA, Kauffman JB (2003) Fire and riparian ecosystems in landscapes of the western USA. Forest
Ecology and Management 178: 61-74
Ehrenfeld JG (2003) Effects of exotic plant invasions on soil nutrient cycling processes. Ecosystems 6: 503-
523
Farnsworth EJ, Meyerson LA (2003) Comparative ecophysiology of four wetland plant species along a
continuum of invasiveness. Wetlands 23: 750-762
Hooper DU, Chapin FS, III, EwelJJ, Hector A, Inchausti P, Lavorel S, LawtonJH, Lodge DM, Loreau M,
Naeem S, Schmid B, Setala H, Symstad AJ, Vandermeer J, Wardle DA (2005) Effects of biodiversity on
ecosystem functioning: A consensus of current knowledge. Ecological Monographs 75: 3-35
Kuebbing SE, Classen AT, Sanders NJ, Simberloff D (2015) Above- and below-ground effects of plant
diversity depend on species origin: an experimental test with multiple invaders. New Phytologist 208
(3):727-735. doi:10.1111/nph.13488
Lesica P (1997) Spread of Phalaris arundinacea adversely impacts the endangered plant Howellia
aquatilis. Great Basin Naturalist 57: 366-368
187
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Lozon JD, Maclsaac HJ (1997) Biological invasions: Are they dependent on disturbance? Environmental
Reviews 5: 131-144
MackJJ, Kentula ME (2010) Metric Similarity in Vegetation-Based Wetland Assessment Methods.
EPA/600/R-10/140. US Environmental Protection Agency, Office of Research and Development,
Washington, DC
Mack RN, Simberloff D, Lonsdale MS, Evans H, Clout M, Bazzaz FA (2000) Biotic invasions: Causes,
epidemiology, global consequences, and control. Ecological Applications 10: 698-710
MageeTK, Ernst TL, Kentula ME, Dwire KA(1999) Floristic comparison of freshwater wetlands in an
urbanizing environment. Wetlands 19: 517-534
Magee TK, Ringold PL, Bollman MA (2008) Alien species importance in native vegetation along wadeable
streams, John Day River basin, Oregon, USA. Plant Ecology 195: 287-307
Magee TK, Ringold PL, Bollman MA, Ernst TL (2010) Index of Alien Impact (IAI): A method for evaluating
alien plant species in native ecosystems. Environmental Management 45: 759-778
Magee TK, Blocksom KA, Herlihy AT, &. Nahlik AM (2019) Characterizing nonnative plants in wetlands
across the conterminous United States. Environmental Monitoring and Assessment 191 (SI): 344, doi:
10.1007/s 10661-019-7317-3. https://link.springer.com/article/10.1007/sl0661-019-7317-3
McKinney ML (2004) Do exotics homogenize or differentiate communities? Roles of sampling and exotic
species richness. Biological Invasions 6: 495-504
Meyerson LA, Mooney HA (2007) Invasive alien species in an era of globalization. Frontiers in Ecology and
the Environments: 199-208
Pysek P, Hulme P, Simberloff D, Bacher S, Blackburn T, Carlton J, Foxcroft L, Genovesi P, Jeschke J, Kuhn I,
Liebhold A, Mandrak N, Meyerson L, Pauchard A, Pergl J, Roy H, Richardson D (2020) Scientists' warning
on invasive alien species. Biological Reviews. doi:10.1111/brv.12627
Randall JM (1996) Weed control for the preservation of biological diversity. Weed Technology 10: 370-
383
Ricciardi A, lacarella JC, Aldridge DC, Blackburn TM, Carlton JT, Catford JA, Dick JTA, Hulme PE, Jeschke
JM, Liebhold AM, Lockwood JL, Maclsaac HJ, Meyerson LA, Pysek P, Richardson DM, Ruiz GM, Simberloff
D, Vila M, Wardle DA (2020) Four priority areas to advance invasion science in the face of rapid
environmental change. Environmental Reviews:l-23. doi:10.1139/er-2020-0088
Ringold PL, Magee TK, Peck DV (2008) Twelve invasive plant taxa in in US western riparian ecosystems.
Journal of North American Benthological Society 27: 949-966
Rooney TP, Wiegmann SM, Rogers DA, Waller DM (2004) Biotic impoverishment and homogenization in
unfragmented forest understory communities. Conservation Biology 18: 787-798
188
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Rowantree K (1991) An assessment of the potential impact of alien invasive vegetation on the
geomorphology of river channels in South Africa. South African Journal of Aquatic Science 17: 28-43
Sala A, Smith SD, Devitt DA (1996) Water use by Tamarix ramosissima and associated phreatophytes.
Ecological Applications 6: 888-898
Seabloom E, Williams J, Slayback D, Stoms D, Viers J, Dobson A (2006) Human impacts, plant invasion, and
imperiled plant species in California. Ecological Applications 16: 1338-1350
Simberloff D (2011) How common are invasion-induced ecosystem impacts? Biological Invasions 13
(5):1255-1268. doi:10.1007/sl0530-011-9956-3
USEPA (2016a) National Wetland Condition Assessment: 2011 Technical Report. EPA-843-R-15-006. .US
Environmental Protection Agency, Washington, DC. https://www.epa.gov/national-aquatic-resource-
surveys/national-wetland-condition-assessment-2011-results
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual (EPA-843-B-21-
002
Vitousek PM, D'Antonio CM, Loope LL, Rejmanek M, Westbrooks R (1997) Introduced species: A
significant component of human-caused global change. New Zealand Journal of Ecology 21: 1-16
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
189
-------
Chapter 11: Human-Mediated Physical Alterations
Physical indicators of disturbance to a wetland site are one of the key categories of data (in addition to
chemical and biological indicators) used in wetland assessment for the NWCA. Six Human-Mediated
Physical Alteration (hereon, "Physical Alteration") indices were developed in addition to two indicators
that integrate scores from all six Physical Alteration indices. Thresholds associated with the six indices and
the two metrics were used for:
• assigning disturbance class and
• indicating stressor condition.
Physical Alteration thresholds used to assign disturbance class are discussed broadly in Chapter 6: (and
specifically in Section 6.3), while thresholds used to indicate stressor condition are provided in Section
11.5 at the end of this chapter. Note that the disturbance class thresholds differ from the stressor
condition thresholds. The methods used to develop the six Physical Alteration indices are discussed in the
following subsections of this chapter.
11.1 Data Collection
In both the 2011 and 2016 NWCAs, two separate protocols (and, thus, two forms) were used in the field
to collect data pertaining to physical disturbances (USEPA 2011a, 2016a); however, the two separate
protocols and some differences between them made both field collection and analysis challenging. The
NWCA Analysis Team decided that for 2021, updated field protocols and forms associated with Physical
Alterations would streamline data collection in the field and produce data more specific to the needs for
analysis and reporting.
The 2021 Physical Alteration field protocol and forms closely reflect the 2011 and 2016 stressor portion
of the Hydrology field protocol and H-l Form and the 2011 and 2016 Buffer field protocol and B-l Form.
Stressors from the previous survey's forms were combined, simplified, and reorganized into 48 Physical
Alteration metrics, equally divided into six indices:
• Vegetation Removal (VEGRMV)
• Vegetation Replacement (VEGRPL)
• Water Addition/Subtraction (WADSUB)
• Flow Obstruction (WOBSTR)
• Soil Hardening (SOHARD)
• Surface Modification (SOMODF)
These indices are collectively referred to as "Human-Mediated Physical Alterations" and indicate human
impacts to the three components that define wetlands: vegetation (Vegetation Removal, Vegetation
Replacement), hydrology (Water Addition/Subtraction, Flow Obstruction), and soils (Soil Hardening,
Surface Modification). Each of the six indices is composed of eight Physical Alteration metrics (Table
11-1).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
190
-------
Table 11-1. Six indices of human-mediated physical alterations and the 48 metrics collected on the P-l Physical Alteration Assessment Area and P-2 Physical
Alteration Buffer Plots Forms.
Physical
Alteration
Index
Physical Alteration Metrics
Forest Clear Cut
>
o
Forest Selective Cut
£ >
Vegetation Damage from Insects
C CH
Herbicide/Pesticide Use
O o
LU
Shrub/Tree Browsing
f0 >
Grass/Forb Grazing
bjO
a)
>
Mowing/Pruning/Clearing
Human-Altered Fire Regime
Abandoned Crop Field/Historical Cultivation
Recent Fallow/Resting Crop Field
o ®
Lawn/Park/Cemetery/Golf Course
~ E £
ro a, DC
Silviculture/Tree Plantation/Orchard/Nursery
^ o n
(D Q. >
> «u —'
CC
Active Row or Field Crop
Range (passively managed)
Pasture (actively managed)
Nonnative Pest Plants
c
Ditch/Channelized Stream (human-made)
.o
Culvert (corrugated pipe, arch, box)
E S-
Point Source/Pipe (effluent, sewer, stormwater)
™ = jq
Tile Drainage/Drain Tiles
(D (/) O
.2 >
Irrigation
Water Withdrawal Pump
~G
¦a
Impervious Surface Input (sheetflow)
<
Human-mediated Shallow Channels (ruts)
Physical
Alteration
Index
Physical Alteration Metrics
Dike/Berm/Levee
C
O
Dam (human-made or beaver-modified structure)
I 2"
Wall/Riprap
I
Trash/Soil/Gravel/Spoil/Organic Debris Heap (human-made)
it
_o
LL.
Road/Railroad/Walkway (raised bed)
Water Level Control Structure
Pond/Retention Basin/Quarry (human-made)
SilvicuItural/AgricuItural Mounding of Soil
Oil/Gas/Utility Wells/Drilling/Pipeline
bjO
Soil Compaction/Pugging/Wallows
C ^
E Q
Non-Paved Trail
-------
The 2021 Physical Alteration field protocol required field crews to observe and record the standardized
set of 48 physical alterations in the entire 5000 m2 AA and in twelve 100 m2 buffer plots (Figure 11-1),
with the same set of 48 physical alterations in the AA and buffer plots (Table 11-1). Each of the 48
physical alterations is clearly defined and associated with examples, which were provided to field crews in
both the 2021 Field Operations Manual and the NWCA 2021 Field App (USEPA 2021a).
• AA Field Protocol - Using the P-l Form, field crews indicated the presence of any of the 48 listed
physical alterations within the entire AA by filling in the most appropriate bubble(s)
corresponding to the estimated percent cover or perceived level of influence in the AA.
• Buffer Field Protocol - Using the P-2 Form, field crews indicated the presence of any of the 48
listed physical alterations within twelve 100-m2 plots located along transects outside of the AA by
filling in the most appropriate bubble(s) next to the plot number.
On both the P-l and P-2 Forms, absence of physical alterations was indicated by a "No Alterations
Present" bubble, showing that they inspected the AA or buffer plot for physical alterations and found
none. On both the P-l and P-2 Forms, the field crews were able write a narrative description of the
physical alterations.
Figure 11-1. The entire AA was evaluated using the P-l Form and 12 buffer plots were evaluated using the P-2
Form.
192
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Logic checks of the data from the field forms identified potential issues that were resolved by the NWCA
Technical Analysis Team. For example, not all the buffer plots were assessed, even though in the final
analysis all plots need to be classified into one of three categories; Not Sampled, Sufficiently Sampled
from Afar, or Sampled. This is important to adjust the final PALT metric scores for missing values.
Keywords from comments on the P-2 Form from sites with Not Sampled or Sufficiently Sampled from Afar
were categorized into a set of six reasons for not sampling the plot or sampling from afar:
• Private Property/Fence/Access permission
• Deep Water
• Road/Railroad/Highway
• Vegetative Barrier
• Other (mostly natural physical barriers or snakes)
• Multiple Barriers
In very few cases, the field crew comments did not support that a buffer plot with data was Sufficiently
Sampled from Afar (as indicated), and the plot in question was set to Not Sampled. Plots designated and
validated as Sampled and Sufficiently Sampled from Afar were considered as valid plot samples for all
future work and no distinction was made between them. For each site, the number of actual sampled
plots in each sample ring position (i.e., buffer plots closest to the AA, buffer plots in the middle, buffer
plots farthest from the AA) so an adjustment for missing values could be made.
11.2 Scoring Each of the Six Physical Alteration Indices
For each site, each of the six Physical Alteration indices (i.e., VEGRMV, VEGRPL, WADSUB, WOBSTR,
SOHARD, and SOMODF) was scored using a proximity-weighted scheme (illustrated in Figure 11-2, Table
11-2), with observations in the AA receiving the highest scores and observations in the furthest buffer
plots from the AA receiving the lowest scores:
• AA Scoring - For each of the six indices, the total points of observed metrics in the AA were
summed with each:
o low influence observation scoring 10 points
o moderate influence observation scoring 25 points
o high influence observation scoring 50 points
• Buffer Scoring - Observed metrics were scored using proximity weighting, with each metric
observed in the:
o inner ring plots scoring 4 points
o middle ring plots scoring 2 points
o outer ring plots scoring 1 point
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
193
-------
o Point
Inner Ring Buffer Plots
Numbers are
Assessment Area
¦
Middle Ring Buffer Plots
the points per
Buffer Area
¦
Outer Ring Buffer Plots
observed metric
Figure 11-2. Physical Alteration plots for the NWCA 2021. Six Physical Alteration indices representing Vegetation
Removal, Vegetation Replacement, Water Addition/Subtraction, Water Obstruction, Soil Hardening, Surface
Modification, each with 8 metrics (i.e., checkboxes from P-l and P-2 forms), are evaluated in the Assessment Area
(AA) and in 12 Buffer Plots. Numbers represent metric scoring (points) associated with observed metrics in the AA
and each Buffer Plot.
Table 11-2. Physical Alteration scoring details for NWCA 2021. Metric scoring is based on proximity to the Point
and the level of influence (in the AA only), with a maximum score of 624 per index for the site.
Points per
Number
Distance from
Maximum Index Score
Location
Observed Metric
of Plots
the Point
Plot Area
per Plot
per Location
10 (low influence)
80
80
Assessment Area
25 (moderate influence)
1
0-40 m
5000 m2(0.5 ha)
200
200
50 (high influence)
400
400
Inner Ring Buffer
4
4
70-80 m
100 m2
32
128
Middle Ring Buffer
2
4
100-110 m
100 m2
16
64
Outer Ring Buffer
1
4
130-140 m
100 m2
8
32
If Physical Alteration data were Not Sampled from the AA, all metrics were set to missing values. If any of
the buffer plots were Not Sampled in the field, the points were redistributed among the number of
Sampled plots within the same ring; for example, if only two of four plots were Sampled in the inner ring,
each metric with observed items would be scored as 8 points (i.e., instead of the 4 points used when all
four plots were Sampled). However, if all four plots were Not Sampled, the ring was assigned 0 points
(i.e., not set to missing).
194
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Lastly, the scores for the six physical indices (VEGRMV, VEGRPL, WADSUB, WOBSTR, SOHARD, and
SOMODF) were calculated individually. The calculation for determining an overall site (PALTSite) score for
any one of the six physical alteration indices is:
PALTSite = PALTaa + PALTbuffe, Eq.l
where for any given site and any of the six indices (listed above), PALTaa is the physical alteration score for
the AA and PALTbuffer is the physical alteration score for the buffer.
11.3 Physical Alteration Screen Scoring (PALT_ANY and PALT_SUM)
Two Physical Alteration screens that integrate scores from all six Physical Alteration indices are calculated
for each site: PALT_ANY and PALT_SUM. Both of these screens are used to set thresholds assigning
disturbance class (Section 6.3), and PALT_SUM (in addition to scores for each index as described in the
previous section) are used to set thresholds for indicating stressor condition.
11.3.1 PALT_ANY
PALT_ANY indicates the maximum degree of Human-Mediated Physical Alterations for any index and is
calculated as the maximum Physical Alteration index score among all six Physical Alteration index scores
for a site.
For any one index at a sampled site, there are only 8 metrics that can be scored within 13 locations (the
AA and 12 buffer plots); therefore, the maximum PALT_ANY score for the site is 624 points:
50 points for high influence * 1 AA *8 metrics
+ 4 points * 4 inner ring buffer plots * 8 metrics
+ 2 points * 4 middle ring buffer plots * 8 metrics
+ 1 point* 4 outer ring plots * 8 metrics
624 maximum points total per index
However, it is implausible that every single metric within an index would occur at the same time in the AA
and in all buffer plots.
11.3.2 PALT_SUM
PALT_SUM is a secondary screen that indicates the cumulative amount of Human-Mediated Physical
Alterations among all indices. It is calculated as the sum of all six Physical Alteration index scores for a
site. This screen was developed to detect instances, e.g., where several metric items were observed, but
the observations are dispersed across several Physical Alteration indices (i.e., no one index has a
particularly high score). Thus, a site may pass the threshold for the PALT_ANY screen and fail the
threshold for the PALT_SUM screen (but not vice versa).
With 624 total points per index, and six indices, the highest possible PALT_SUM score for a site is 3,744.
However, it is implausible that every single metric within an index would occur at the same time in the AA
and in all buffer plots, much less across all indices.
195
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
11.4 Physical Alteration Stressor Condition Thresholds
Like other National Aquatic Resource Survey (NARS) assessments, the NWCA data was used to identify
connections between the presence of indicators of stress and ecological condition. Anthropogenic
stressors act to degrade ecological condition, and consequently, evaluation of indicators of stress is an
important component of an assessment method (Fennessy et al. 2007). Using physical, chemical, and
human-health indicators of stress, the NWCA analysis examined a variety of stressor data to detect
factors likely affecting wetland condition. The use of stressor data is consistent with current approaches
to assess wetlands and recognizes the connection between the presence of stressors and wetland
condition. For example, rapid assessment methods have been developed which use only stressors as
indicators of condition (e.g., the Delaware Rapid Assessment Method (Jacobs 2007)) and models
comprising an HGM assessment (a Level 3, intensive assessment) use stressors as variables (e.g.,
Whigham et al. 2007, Wardrop et al. 2007). The data sources for the indicators of stressor condition used
in the NWCA analysis were primarily from field observations and soil and water chemistry samples
collected from the Assessment Area (AA) and its buffer at each sampled site.
Seven physical indicators of stressor condition are reported for the 2016 NWCA:
• Vegetation Removal (VEGRMV),
• Vegetation Replacement (VEGRPL),
• Water Addition/Subtraction (WADSUB),
• Flow Obstruction (WOBSTR),
• Soil Hardening (SOHARD),
• Surface Modification (SOMODF), and
• Physical Alterations (PALT_SUM).
In contrast to the Disturbance Gradient, six individual Physical Alteration indices are used instead of the
PALT_ANY screen to indicate stressor condition. The reasoning for this decision to use the six individual
indices was to provide condition extent and relative and attributable risk associated with each of these
specific indicators.
For the six main Physical Alteration indicators (all indices except PALT_SUM), each site was assigned to
"good", "fair", or "poor" stressor condition based on thresholds for each indicator. The same national
thresholds were used for the six main indicators, with sites scoring:
• 0 points assigned to good stressor condition,
• >25 points assigned to poor stressor condition, and
• >0 and < 25 points (i.e., everything between good and poor) assigned to fair stressor condition.
Thresholds were chosen based on common sense for the good condition threshold (i.e., the expectation
for a good condition site is to have no observed physical alterations), and best professional judgement for
the poor condition threshold. Analysts chose > 25 because it conceptually works with the scoring in the
AA and buffer: i.e., 2 low impact stressors (10 points each) in AA is still fair, but 1 moderate or high impact
stressor (25 or 50 points) in AA is high. Likewise, a stressor could be observed in several of the buffer
plots (and not in the AA) and still be in fair condition. But a low impact stressor in the AA and many in the
buffer would possibly shift the site from fair to poor.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
196
-------
In addition, the cumulative Physical Alteration indicator (PALT_SUM) was calculated by summing the
scores of all six Physical Alteration index scores for a site. This was developed to detect instances where
several metric items were observed, but the observations are dispersed across several Physical Alteration
indices (i.e., no one index has a particularly high score). Thus, a site may pass the threshold for the
individual index and fail the threshold for the cumulative index (but not vice versa).
For the cumulative Physical Alteration indicator (PALT_SUM), the thresholds were set to:
• 0 points assigned to good stressor condition,
• >50 points assigned to poor stressor condition, and
• >0 and < 50 points (i.e., everything between good and poor) assigned to fair stressor condition.
A poor condition site means that there were 2 moderate stressors in the AA or 1 high influence stressor in
the AA. 1 moderate stressor in the AA and several (enough to add up to 25 points) in the buffer would
also be categorized as poor.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
197
-------
11.5 Literature Cited
Fennessy MS, Jacobs AD, Kentula ME (2007) An evaluation of rapid methods for assessing the ecological
condition of wetlands. Wetlands 27: 543-560
Jacobs AD (2007) Delaware Rapid Assessment Procedure Version 4.1. Delaware Department of Natural
Resources and Environmental Control, Dover, DE
USEPA (2011a) National Wetland Condition Assessment: Field Operations Manual. EPA-843-R10-001. US
Environmental Protection Agency, Washington, DC
USEPA (2016a) National Wetland Condition Assessment 2016: Field Operations Manual. EPA-843-R-15-
007. US Environmental Protection Agency, Washington DC
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual (EPA-843-B-21-
002
Wardrop DH, Kentula ME, Jensen SF, Stevens Jr. DL, Hychka KC, Brooks RP (2007) Assessment of wetlands
in the Upper Juniata watershed in Pennsylvania, USA using the hydrogeomorphic approach. Wetlands 27:
432-445
Whigham DF, Jacobs AD, Weller DE, Jordan TE, Kentula ME, Jensen SF, Stevens Jr. DL (2007) Combining
HGM and EMAP procedures to assess wetlands at the watershed scale - Status of flats and non-tidal
riverine wetlands in the Nanticoke River Watershed, Delaware and Maryland (USA). Wetlands 27: 462-
478
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
198
-------
Chapter 12: Soil Heavy Metals
Soil heavy metal results are not yet available from the laboratory. This chapter will be updated when
results are ready and analysis of the soil data collected in NWCA has been completed.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
199
-------
National Wetland Condition Assessment: 2021 Technical Support Document
Dec 2024
-------
Chapter 13: Water Chemistry
Chemical indicators of disturbance to a wetland site are one of the key categories of data (in addition to
physical and biological indicators).. In the NWCA 2016, two water chemistry parameters, total nitrogen
(TN) and total phosphorus (TP) concentrations, were developed as core indicators of stressor condition.
In order to use TN and TP as indicators of stressor condition, several distinct procedures needed to be
completed to provide the basis for setting TN and TP stressor condition thresholds and include:
• Describing the population of wetlands sampled for water chemistry (Section 13.3);
• Developing Physical-Alterations-Possibly-Affecting-Chemicals (CALT) indices for use in screening
sites to establish the disturbance gradient for sites sampled for water chemistry;
• Using two CALT indices and four landscape metrics and their associated disturbance thresholds to
screen sites and assign disturbance classes (i.e., "least disturbed", "intermediate disturbed", and
"most disturbed"); and, finally,
• Calculating "good", "fair", and "poor" thresholds for stressor condition using the 75th and 95th
percentiles of TN and TP concentrations among least-disturbed sites sampled for water
chemistry.
These procedures are discussed in this chapter, and were the basis for reporting the extent of TN and TP
stressor conditions for wetlands sampled for water chemistry in the NWCA 2016 and 2021 reports.
Section 13.1 through 13.4 describe the data collection, validation, and development of TN and TP
thresholds for stressor condition using the NWCA 2016 water chemistry data. Section 13.5 describes the
application of the TN and TP thresholds to the NWCA 2021 water chemistry data.
13.1 Data Collection
Water chemistry samples were collected at all wetlands having sufficient sampleable surface water within
the 0.5 ha assessment area (AA) during the sampling visit. Because surface water was required to be
within the AA, not all sites yielded a water sample - even when surface water was present elsewhere in
the wetland. Furthermore, some wetlands lacked surface water entirely during the sampling visit. Sixty-
four percent of probability and handpicked sites across both Visit 1 and Visit 2 yielded a water sample in
2016. The percentage of 2016 sites with water chemistry samples is approximately 10% higher than in the
2011 NWCA, largely attributed to the removal of the 2011 sampling location water-depth-minimum of 15
cm.
Laboratory analyses were conducted per methods detailed in 2016 Field Operations Manual (USEPA
2016a) and in Table 13-1. In summary, water chemistry sampling consisted of using a dipper to fill 1) a 1L
bottle that was filtered on-site for later chlorophyll-a analysis, and 2) a 1L cubitainer for laboratory
analysis of other water chemistry parameters. The chlorophyll filters and cubitainers were chilled
immediately and express-shipped to the USEPA Pacific Ecological Systems Division (PESD) in Corvallis,
Oregon for analyses.
In addition to the analytes measured in the lab (Table 13-1), conductivity and pH were measured in the
field at some sites (at the field crew's discretion). While most analytes were measured in both the 2011
and 2016 surveys, four analytes were added to the 2016 analysis: turbidity, DOC, chloride, and sulfate.
Chloride and sulfate, important indicators of anthropogenic disturbance (e.g., water softeners, fertilizers,
and road salt for chloride (Herlihy et al. 1998) and mine influences for sulfate (Herlihy et al. 1990)), were
only measured in freshwater samples. There is no expectation that chloride and sulfate concentrations in
201
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
saltwater would be informative of anthropogenic impacts (i.e., concentrations would reflect the saltwater
influence).
Table 13-1. Water chemistry analytes measured in the laboratory, with their associated units and a summary of
methods.
Analyte
Units
Summary of Method
Conductivity
|iS/cm at 25°C
Electrolytic
pH (laboratory)
Standard (Std) Units
Automated with autotitrator and combination pH electrode
or manual electrolytic analysis
Turbidity
Nephelometric Turbidity
Units (NTU)
Automated nephelometric analysis or manual turbidmetric
analysis (high turbidity samples)
Dissolved Organic
Carbon (DOC)
mg-C/L
UV promoted persulfate oxidation to C02 with infrared
detection
Ammonia (NH3)
mg-N/L
FIA automated colorimetric (with use of salicylate,
dichloroisocyan urate)
Nitrate-Nitrite (N03-N02)
mg-N/L
Ion chromatography (freshwater samples) or FIA
automated colorimetric (cadmium reduction for brackish or
freshwater samples)
Total Nitrogen (TN)
mg/L
Persulfate digestion followed by FIA automated
colorimetric analysis
Total Phosphorus (TP)
mg-P/L
Persulfate digestion followed by FIA automated
colorimetric analysis
Sulfate (S04)
mg-S04/L
Ion Chromatography (freshwater samples only)
Chloride (CI)
mg-CI/L
Ion Chromatography (freshwater samples only)
Chlorophyll-a
M-g/L
90% acetone exraction followed by fluorometry analysis
13.2 Data Validation
Data validation refers to the process of checking for completeness and repeatability the data, which
begins upon receiving the data from participating laboratories through the assembly of data into results
files. Validation is especially important for water chemistry data because samples were processed by
multiple state and regional laboratories across the US. Data validation was completed for all water
chemistry parameters using completeness-checking, repeatability-checking, and evaluation of cross-visit
repeatability. Details about how each of these methods were applied to the data are discussed in the
following paragraphs.
202
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Completeness-checking refers to checking and addressing any missing values or any data values that
should be set to missing because of documented collection or analysis concerns. Water chemistry
analytes whose values were flagged as being below the laboratory's minimum detection limit (MDL) were
generally assigned a value of half that detection limit. Per Hornung and Reed (1990), the practice of using
half the MDL more accurately preserves the data distribution properties than alternatives, such as setting
below-detection values to zero. An exception to this general rule was made for certain chlorophyll-a and
ammonia samples. These few samples had high detection limits due to either 1) the amount of water
filtered in the field or 2) the amount of dilution that occurred in the laboratory before analysis to get the
sample within instrument range. To avoid over-inferring concentration values these samples that were
poorly characterized by such high detection limits, flagged samples with MDLs above 2.0 |ag/L for
chlorophyll-a and 0.03 mg-N/L for ammonia were set to "missing" (i.e., "NA" in the database).
Data for TN and TP were complete across the dataset (i.e., no missing values), as were data for ammonia,
nitrate/nitrite, conductivity, pH, and turbidity. Chloride and sulfate data, which are associated exclusively
with freshwater, were not analyzed (i.e., missing) from 55 saltwater sites identified by high conductivity
levels. Several DOC values were missing because one laboratory erroneously analyzed total organic
carbon (TOC). Chlorophyll-a values were laboratory-reported as "missing" from four sites due to problems
with filter type or filter volume, and an additional 27 sites were set to "missing" due to flagged samples
with MDLs above 2.0 |ag/L. While the dataset started as complete, 26 ammonia values were set to
"missing" due to flagged samples with MDLs above 0.03 mg-N/L, and five cases were missing because the
ammonia concentration of the sample exceeded that of TN (indicating measurement error).
Repeatability-checking included the comparison of analyte values between Visit 1 and Visit 2 (for the
approximately 10% of sites where a second visit was done), and comparison of any field measurements
for conductivity and pH to the corresponding laboratory measurements. The field versus laboratory
comparisons revealed several cases of conductivity being recorded in the wrong units in the field (e.g.,
milliSiemens rather than microSiemens per centimeter), likely because of limitations on the field meter
display. Once these were corrected, the Pearson correlation between field-measured and lab-measured
conductivity was extremely high (r = 0.99), confirming that conductivity is consistent between laboratory
and field measurements. On the other hand, there are consistent differences in laboratory-measured pH
and field-measured pH (r = 0.72) - likely driven by varying degrees of carbon dioxide (C02) saturation.
Parallel to findings from the 2011 survey (USEPA 2016c), laboratory-measured pH values tended to be
higher than those measured in the field for acidic waters (i.e., pH < 7.0), while laboratory-measured pH
values tended to be lower than those measured in the field for alkaline waters (i.e., pH > 7.0).
Cross-visit repeatability can be assessed directly by analyzing the correlation of values between visits to
the same site within the same year (i.e., Visit 1 compared to Visit 2). However, the interpretation of cross-
visit repeatability is affected by the rate at which below-detection (i.e., MDL) values occur for any given
analyte. Abundant data below the MDL (e.g., NH4 and N03) results in the same low below detection limit
values, leaving few data to correlate.
Comparing the variance associated with a sampling site (signal) to the variance associated with repeated
visits to the same site (noise) results in the Signal-to-Noise Ratio (S:N) (Kaufmann et al. 1999, 2014),
which is described in detail in Chapter 8:, Section 8.5.2. All sites are included in the signal, whereas only
revisit sites contribute to the noise component. S:N is a useful for discerning environmentally-significant
patterns for an analyte against the background of its typical variability. Analytes with high S:N are more
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
likely to show consistent responses, and S:N values < 1 indicate that sampling a site twice yields as much
or more variability as sampling two different sites (Stoddard et al. 2008).
Considering all sites sampled in the 2016 NWCA, only nitrate-nitrite, chloride, and TN had S:N < 3 (Table
13-2), indicating that these analytes had high within-site variability. Chloride (and to a lesser extent
sulfate), which were not measured in saltwater sites but may have been measured in brackish sites
(discussed previously in Section 13.1), had low S:N for all 2016 sites as a result of high between-visit
variability in estuarine sites. However, when S:N is calculated for inland (i.e., freshwater) sites, the ratio
for both chloride and sulfate increased to > 10.
Table 13-2. Variability and repeatability of water chemistry analytes measured in the 2016 NWCA, including below-
detection rates for all 2016 NWCA sites (Visit 1 and Visit 2, probability and handpicked), cross-visit correlations
based on the 61 revisit sites, and Signal-to-Noise ratios (S:N) for all sites and inland (freshwater) sites.
Analyte
Below-Detection
Rate
Cross-Visit
Pearson
Correlation (r)
S:N
S:N
(All 2016 Sites)
(2016 Inland Sites)
Conductivity
None
0.97
20.9
29.7
pH (laboratory)
NA
0.88
15.9
18.3
Turbidity
0.3%
0.58
40.1
37.6
Dissolved Organic Carbon (DOC)
0.2%
0.87
6.67
7.11
Ammonia (NH3)
37.1%
0.10
3.94
4.30
Nitrate-Nitrite (N03-N02)
33.6%
0.28
1.97
3.11
Total Nitrogen (TN)
None
0.39
1.98
1.55
Total Phosphorus (TP)
0.3%
0.85
17.6
14.7
Sulfate (S04)
2.7%
0.94
3.09
12.9
Chloride (CI)
0.3%
0.99
1.40
10.3
Chlorophyll-a
7.3%
0.62
11.9
13.1
13.3 Establishing a Disturbance Gradient for Sites Sampled for Water
Chemistry
The wetland population represented by water chemistry is a subset of the larger NWCA wetland
population; 56% and 65% of the wetlands in 2011 and 2016, respectively, sampled across both Visit 1 and
Visit 2 had sufficient surface water to collect and analyze. Thus, water chemistry data were excluded from
the generation of the disturbance gradient used to identify abiotic and final least- and most-disturbed
sites (i.e., ABIOTIC_REF_NWCA and REF_NWCA), discussed in Chapter 6:.
However, in order to develop chemical indicators of stressor condition based on TN and TP measured in
the water column (presented later in Section 13.4), it is necessary to create a specially-defined
disturbance Rradient for the subset of sites that were sampled for water chemistry. To establish a water
chemistry disturbance gradient, all 1,198 unique probability and handpicked sites across both the 2011
NWCA and the 2016 NWCA that were sampled for water chemistry (Table 13-3) were used. The general
process for setting least-disturbed and most-disturbed thresholds, and for assigning disturbance class is
discussed in Chapter 6:, Section 6.2.2 through Section 6.2.4. Here, the process used for assigning least-
disturbed and most-disturbed water chemistry sites is described, beginning with the development of
indices used to develop least- and most-disturbed thresholds.
204
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 13-3. The number of Visit 1 (VI) probability and handpicked sites sampled for water chemistry in 2011 and
2016, with their totals. Additionally, the numbers of resampled sites with water chemistry data are reported in
paratheses to indicate that these are subtracted from the subtotals above. The total number of unique probability
and handpicked sites with water chemistry data are reported with the final number of Index Visit sites (in the red
cell) used in the establishment of the water chemistry disturbance gradient. Note that this table does not include
the 51 Visit 2 sites with water chemistry sampled in 2011 and 64 Visit 2 sites with water chemistry sampled in
2016, which are only used to calculate Signal-to-Noise ratios.
SURVEY YEAR
VI PROBABILITY
WITH WATER
CHEMISTRY
(n-sites)
HANDPICKED
WITH WATER
CHEMISTRY
(n-sites)
TOTAL
2011 NWCA
531
86
617
2016 NWCA
611
64
675
SUBTOTAL
1142
150
1292
2011 Sites with Water Chemistry Resampled in 2016
(94)
(0)
(94)
TOTAL UNIQUE SITES WITH WATER CHEMISTRY
1048
150
1198
13.3.1 Development of Physical-Alterations-Possibly-Affecting-Chemicals (CALT) Indices
Three indices, collectively referred to as the Physical-Alterations-Possibly-Affecting-Chemicals (CALT)
indices, were developed for use in screening sites to establish the disturbance gradient for sites sampled
for water chemistry:
• the CALT_NUT index, alterations thought likely to affect nutrient levels,
• the CALT_SED index, alterations thought likely to affect suspended sediment levels, and
• the CALT_SAL index, alterations thought likely to affect salinity levels.
These three indices were based on observational data associated with the stressor check lists (hereon
referred to as "Items") on the H-l and B-l Forms (see Table 11-1). Best professional judgement (BPJ) was
used to evaluate if and how each H-l and B-l Form Item might affect nutrients, suspended sediments,
and salinity at a site. Write-in "others" were not considered and were therefore excluded from this
analysis. Based on this evaluation, a subset of the H-l and B-l Form Items were assigned to one, two, or
all three of the Chemical-Response-to-Physical-Alteration indices (Table 13-4). Note that not all metrics
were assigned to one of the CALT indices (hence, "subset").
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
205
-------
Table 13-4. Subset of Physical Alteration metrics (defined in Chapter 11:, Section 11.2) assigned to the Physical-
Alterations-Possibly-Affecting-Nutrient (CALT_NUT), Physical-Alterations-Possibly-Affecting-Suspended Sediments
(CALT_SED), and Physical-Alterations-Possibly-Affecting-Salinity (CALT_SAL) indices. "X" indicates that the Physical
Alteration metric was included the CALT index. Note that not all Physical Alteration metrics were assigned to a
CALT index. Write-in "others" from the H-l and B-l Forms were not considered and are therefore excluded from
the list of Form Items.
Form
Form Items
2016 Parameter Name
Nutrients
(CALT_NUT)
Suspended Sediments
(CALT_SED)
Salinity
(CALT_SAL)
B-l
Forest Clear Cut
HAB CLEAR-CUT
X
X
B-l
Forest Selective Cut
HAB SELECTIVE CUT
X
B-l
Tree Canopy Herbivory (insect)
HABJHERBIVORY
B-l
Herbicide/Pesticide Use
HAB HERBICIDE PESTICIDE
B-l
Shrub Layer Browsed (wild or domestic)
HAB SHRUB
B-l
Highly Grazed Grasses (overall <3" high)
HAB GRAZED
X
X
B-l
Mowing/Shrub Cutting
HAB_MOWING
B-l
Recently Burned Forest (canopy)
HAB FOREST BURNED
X
B-l
Recently Burned Grassland (blackened)
HAB_G RASS_BU RN E D
X
B-l
Fallow Field (old - grass, shrubs, trees)
AGR FALLOW OLD
B-l
Fallow Field (recent - resting row crop field)
AGR FALLOW RECENT
X
B-l
Golf Course
RES GOLF
X
X
X
B-l
Lawn/Park
RES_LAWN
X
X
B-l
Orchard/Nursery
AGR ORCHARD
X
B-l
Silviculture/Tree Plantation
HAB PLANTATION
B-l
Row Crops - Tilling
AGR_ROW
X
X
X
B-l
Range
AGR RANGE
X
B-l
Pasture/Hay
AGR_PASTURE
X
H-l
Culverts & Ditching: Ditches
DITCH_PRESENT
H-l
Culverts & Ditching: Channelized Streams
CHANNELIZED PRESENT
H-l
Culverts & Ditching: Corrugated Pipe
CORR PRESENT
H-l
Culverts & Ditching: Box
BOX_PRESENT
H-l
Pipes: Sewer Outfall
SEWER PRESENT
X
X
H-l
Pipes: Standpipe Outflow
STANDPIPE PRESENT
H-l
Field Drainage Tiling
TILING_PRESENT
H-l
Pumps: Irrigation
IRRIGATION PRESENT
X
H-l
Pumps: Other
PUMP OTHER PRESENT
X
H-l
Pumps: Water Supply
WAT SUPPLY PRESENT
H-l
Shallow Channels: Vehicle Ruts
RUTS_PRESENT
X
H-l
Shallow Channels: Abandoned Road
ABANDONED PRESENT
X
H-l
Shallow Channels: Eroded Foot Paths
PATHS_PRESENT
X
H-l
Shallow Channels: Trails
TRAILS_PRESENT
X
H-l
Shallow Channels: Animal Trampling
ANTRAMP PRESENT
X
X
B-l
Ditches, Channelization
HYD DITCH
B-l
Inlets, Outlets
HYDJNLETS
B-l
Point Source/Pipe (effluent or stormwater)
HYD PIPE
B-l
Drain Tiling
AGR TILING
X
X
B-l
Irrigation
AGR IRRIGATION
X
X
B-l
Impervious Surface Input (sheetflow)
HYDJMPERVIOUS
X
206
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Form
Form Items
2016 Parameter Name
Nutrients
(CALT_NUT)
Suspended Sediments
(CALT_SED)
Salinity
(CALT_SAL)
H-l
Damming Features: Dikes
DIKES_PRESENT
H-l
Damming Features: Berms
BERMS PRESENT
H-l
Damming Features: Dams
DAMS PRESENT
H-l
Damming Features: Roads (all types)
ROADS_PRESENT
X
X
H-l
Damming Features: Railroad Bed
RRBED PRESENT
B-l
Dike/Dam/Road/RR Bed (impede flow)
HYD DDRR
B-l
Wall/Riprap
HYD_WALL
B-l
Fill/Spoil Banks
HYD FILL
X
B-l
Water Level Control Structure
HYD_WATER
H-l
Impervious Surfaces: Compacted non-paved (on 2016 H-l Form only)
Impervious Surfaces: Roads (on 2011 H-l Form only)
IMPER_ROADS_PRESENT
X
H-l
Impervious Surfaces: Asphalt
IMPER ASPHALT PRESENT
X
H-l
Impervious Surfaces: Concrete
IMPER_CONCRETE_PRESENT
X
B-l
Oil/Gas Wells/Drilling
IND OIL GAS
X
X
B-l
Confined Animal Feeding
AGR ANIMAL
X
X
X
B-l
Dairy (on 2011 B-l Form only)
Livestock or Domesticated Animals (on 2016 B-l Form Only)
AGR_DAIRY
X
X
X
B-l
Soil Compaction (animal or human)
HAB SOIL
X
B-l
Trails
HAB TRAILS
X
B-l
Off road Vehicle Damage
HAB_ORV
X
B-l
Road (paved or unpaved) (on 2016 B-l Form only)
Road - Gravel (on 2011 B-l Form only)
Road - Two Lane (on 2011 B-l Form only)
Road - Four Lane (on 2011 B-l Form only)
RES_ROAD
X
B-l
Parking Lot/Pavement
RES LOT
X
X
B-l
Rural Residential
AGR_RURAL
X
B-l
Suburban Residential
RES RES
X
X
B-l
Urban/Multifamily
RES URBAN
X
B-l
Power Line
RES_POWER
H-l
Excavation/Dredging
EXCAVATION PRESENT
X
H-l
Recent Sedimentation
SEDIMENT PRESENT
X
B-l
Dumping
RES_DUMPING
B-l
Trash
RES TRASH
B-l
Landfill
RES LANDFILL
X
X
X
B-l
Excavation, Dredging
HYD EXCAVATION
X
B-l
Gravel Pit
AG R_G RAVEL
X
X
B-l
Mine (surface/underground)
IND MINING
X
X
B-l
Freshly Deposited Sediment (unvegetated)
HYD SEDIMENT
X
B-l
Soil Erosion/Deposition (from wind, water, or overuse)
HAB_EROSION
X
B-l
Soil Loss/Root Exposure
HYD_SOIL
X
B-l
Military
IND_MILITARY
X
X
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
207
-------
For each site, each of the three Physical-Alterations-Possibly-Affecting-Chemicals indices (i.e., CALT_NUT,
CALT_SED, and CALT_SAL) was scored by simply tallying the number of B-l and H-l Items observed and
weighting each tally using the same proximity-weighted scheme used for the Physical Alteration indices.
Observations in the AA received the highest scores (25 points for each tally) and observations in the
buffer plots received increasingly lower weighted scores with distance from the AA (inner-ring buffer
plots = 4 points per tally, middle-ring buffer plots = 2 points per tally, and outer-ring buffer plots = 1 point
per tally). Detailed scoring protocol and a scoring illustration can be found in Chapter 11:, Section 11.3
and Figure 11-2.
Site CALT_NUT, CALT_SED, and CALT_SAL scores range from 0 points (no items from FH-1 or B-l Forms
observed) to almost 300 points, although few sites scored over 100 points.
TP, and to a lesser extent TN were correlated to CALT_NUT and CALT_SAL and these indices were used to
help define the disturbance gradient screen. The third CALT index, the Physical-Alterations-Possibly-
Affecting-Suspended Sediments (CALT_SED) index, was not included as a disturbance gradient screen
(i.e., it was excluded from further use) because analysis revealed that it was not sufficiently related to
nutrients or turbidity. Turbidity, a measure of the degree to which a beam of light passed through a water
sample is attenuated by particulates in that water, is the one NWCA 2016 water chemistry measurement
that might be expected to respond to sediment loading, but the ability to see such a response can be
weakened by 1) the fact that turbidity also reflects the concentration of plankton algae in the water
column, and 2) unless they are derived from very fine-grained sediments (e.g., clays), sediments loaded to
wetlands settle out of the water column rather quickly.
13.3.2 Screens and Thresholds for Sites Sampled for Water Chemistry
Six physical and landscape screens were used to identify least-disturbed, intermediate-disturbed, and
most-disturbed sites sampled for water chemistry. These screens include two Physical-Alterations-
Possibly-Affecting-Chemicals (CALT) indices and four landscape metrics:
• the Physical-Alterations-Possibly-Affecting-Nutrients (CALT_NUT) index,
• the Physical-Alterations-Possibly-Affecting-Salinity (CALT_SAL) index,
• the Percent Agriculture in the 1000-m buffer surrounding the AA,
• the Percent Developed in the 1000-m buffer surrounding the AA,
• the Percent Agriculture in the HUC-12 in which the AA was located, and
• the Percent Developed in the HUC-12 in which the AA was located.
Land cover derived from 30-m resolution 2011 and 2016 rasters (depending on the NWCA collection year
of the data being screened) of the 2016 National Land Cover Database (NLCD, Dewitz 2019) were used to
calculate the extent of agriculture and developed land cover for the 1000-m buffer surrounding the AA
and the US Geological Survey (USGS) 12-digit Hydrologic Unit Code (HUC-12) in which the AA was located.
The extent (percent) of agriculture encompasses Planted/Cultivated Classes and includes NLCD Values 81
and 82.8 The extent (percent) of developed encompasses the Developed Class and includes NLCD Values
8 Value 81 = Pasture/Hay - areas of grasses, legumes, or grass-legume mixtures planted for livestock grazing or the
production of seed or hay crops, typically on a perennial cycle. Pasture/hay vegetation accounts for greater than
20% of total vegetation. Value 82 = Cultivated Crops - areas used for the production of annual crops, such as corn,
soybeans, vegetables, tobacco, and cotton, and also perennial woody crops such as orchards and vineyards. Crop
vegetation accounts for greater than 20% of total vegetation. This class also includes all land being actively tilled.
(https://www.mrlc.gov/data/legends/national-land-cover-database-class-legend-and-description)
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
21, 22, 23, and 24.9 Land cover at the 1000-m buffer scale and at the HUC-12 scale were highly correlated
(r = 0.7) but different enough to filter different sites, so both screens for both scales were used to help
define the disturbance gradient for sites sampled for water chemistry.
National thresholds (i.e., the same thresholds regardless of region) for "least disturbed" and "most
disturbed" were used for all six screens and are reported in Table 13-5. Sites that passed all six screens
were considered "least disturbed" while sites that exceeded any one of the six most-disturbed thresholds
were considered "most disturbed". All other sites were assigned to the intermediate disturbance class.
Table 13-5. Six water chemistry screens and their least-disturbed and most-disturbed thresholds used to assign
disturbance class to each site sampled for water chemistry.
Water Chemistry Screen
Least-Disturbed
Thresholds
Most-Disturbed
Thresholds
Physical-Alterations-Possibly-Affecting-Nutrients (CALT_NUT)
< 5 points
> 50 points
Physical-Alterations-Possibly-Affecting-Salinity (CALT_SAL)
< 5 points
> 50 points
% Agriculture in 1000-m buffer
< 5%
> 50%
% Developed in 1000-m buffer
< 5%
> 50%
% Agriculture in HUC-12
< 5%
> 50%
% Developed in HUC-12
< 5%
> 50%
A summary of the number of sites within each water chemistry disturbance class are reported by region
(RPT_UNIT_5) in Table 13-6. There were 1,198 unique NWCA sites (see Table 13-3) that had measured
water chemistry with roughly equal sample sizes among the five reporting units. However, there were
only six least-disturbed sites in the Plains (PLN). Even though so few sites are not ideal for analysis, the
least-disturbed threshold would have needed to be so severely relaxed to gain the optimal 30-50 least-
disturbed sites for PLN, that least-disturbed and most-disturbed thresholds would have been almost
equivalent. Three sites that lacked CALT scores were assigned as "unknown".
9 Value 21 = Developed, Open Space - areas with a mixture of some constructed materials, but mostly vegetation
in the form of lawn grasses. Impervious surfaces account for less than 20% of total cover. These areas most
commonly include large-lot single-family housing units, parks, golf courses, and vegetation planted in developed
settings for recreation, erosion control, or aesthetic purposes. Value 22 = Developed, Low Intensity - areas with a
mixture of constructed materials and vegetation. Impervious surfaces account for 20% to 49% percent of total
cover. These areas most commonly include single-family housing units. Value 23 = Developed, Medium Intensity -
areas with a mixture of constructed materials and vegetation. Impervious surfaces account for 50% to 79% of the
total cover. These areas most commonly include single-family housing units. Value 24 = Developed, High Intensity
- highly developed areas where people reside or work in high numbers. Examples include apartment complexes,
row houses and commercial/industrial. Impervious surfaces account for 80% to 100% of the total cover.
(https://www.mrlc.gov/data/legends/national-land-cover-database-class-legend-and-description)
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Table 13-6. n-sites sampled for water chemistry, presented by disturbance class assignments (unpublished)
reported by region (RPT_UNIT_5) for Visit 1, Index Visit 2011 and 2016 sites.
Least
Intermediate
Most
Regional
Region
Disturbed (L)
Disturbed (1)
Disturbed (M)
Unknown (?)
Totals
Tidal Saline (TDL)
92
204
32
1
329
Inland Coastal Plains (ICP)
31
117
40
1
189
E. Mts & Upper Midwest (EMU)
59
138
32
1
230
Plains (PLN)
6
65
142
0
213
West (WST)
56
91
90
0
237
National Totals
244
615
336
3
1198
13.3.3 Evaluation of the Disturbance Gradient for Sites Sampled for Water Chemistry
The disturbance gradient for sites sampled for water chemistry was developed to support the
development of TN and TP as indicators of stressor condition. Least-disturbed sites will serve as the
foundation for defining TN and TP stressor condition thresholds (presented in the next section, Section
13.4), thus, it is imperative that least-disturbed sites are distinguished from most-disturbed sites in both
the TN and TP data.
Using LoglO-transformed TP and TN, t-tests performed on national data (i.e., unique 2011 and 2016 Visit
1 sites) showed that distinction of least-disturbed from most-disturbed sites was highly significant (t > 11,
p < 0.001). Figure 13-1 illustrates this distinction among five regions (RPT_UNIT_5). However, statistical
analyses showed that there were no differences between least- and most-disturbed sites in Tidal Saline
(TDL) wetlands, and there were not enough least-disturbed sites in the Plains (PLN) to reach any statistical
conclusions. Significant differences (t = 4 to 8, p < 0.001) were found for the Inland Coastal Plains (ICP),
Eastern Mountains and Upper Midwest (EMU), and West (WST). TP differences were generally stronger
than TN differences (Figure 13-1).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
210
-------
txo
E
o
ro
QJ
u
c
o
u
10.0-
1.0
0.1
a) Total Nitrogen (TN)
I I Least-Disturbed
|~1 Most-Disturbed
TDL
ICP EMU PLN WST
10000
O
•¦P
fB
QJ
u
c
o
u
Q_
1000
100
10
b) Total Phosphorus (TP)
l~l Least-Disturbed
TDL ICP EMU PLN WST
Figure 13-1. Box and whisker plots showing differences between least-disturbed (blue) and most-disturbed (red)
(unique 2011 and 2016 Visit 1) sites among five regions (RPT_UNIT_5) for a) total nitrogen (TN) and b) total
phosphorus (TP). TDL = Tidal Saline, ICP = Inland Coastal Plains, EMU = Eastern Mountains & Upper Midwest, PLN :
Plains, and WST = West.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
13.4 TN and TP Stressor Condition Thresholds
Like other National Aquatic Resource Survey (NARS) assessments, the NWCA data was used to identify
connections between the presence of indicators of stress and ecological condition. Anthropogenic
stressors act to degrade ecological condition, and consequently, evaluation of indicators of stress is an
important component of an assessment method (Fennessy et al. 2007). Using physical, chemical, and
human-health indicators of stress, the NWCA analysis examined a variety of stressor data to detect
factors likely affecting wetland condition. The use of stressor data is consistent with current approaches
to assess wetlands and recognizes the connection between the presence of stressors and wetland
condition. For example, rapid assessment methods have been developed which use only stressors as
indicators of condition (e.g., the Delaware Rapid Assessment Method (Jacobs 2007)) and models
comprising an HGM assessment (a Level 3, intensive assessment) use stressors as variables (e.g.,
Whigham et al. 2007, Wardrop et al. 2007). The data sources for the indicators of stressor condition used
in the NWCA analysis were primarily from field observations and soil and water chemistry samples
collected from the Assessment Area (AA) and its buffer at each sampled site.
Two water chemistry indicators of stressor condition are reported for the 2016 NWCA: 1) total nitrogen
(TN) and 2) total phosphrous (TP) concentrations measured in the water column of sampled sites with
water. Because TN and TP are highly variable in wetlands depending on the wetland type, hydrology, and
other defining characteristics of wetlands that influence nutrient cycling, there is no concurrence in the
literature about expected "reference" concentrations of TN and TP. Thus, thresholds for water column TN
and TP were developed using the same percentile approach that is used by NARS (e.g., Paulsen et al.
2008, USEPA 2016d).
First, subpopulations for which thresholds should be developed needed to be determined. This was
completed by evaluating concentrations of TN and TP in least-disturbed sites sampled for water chemistry
(defined in the previous Section 13.3 and in Table 13-6) across regional subpopulations (specifically, Five
Reporting Units (RPT_UNIT_5)). The results of these evaluations, illustrated in Figure 13-1, indicated that
there were no significant differences in TN concentrations across the least-disturbed sites among the Five
Reporting Units (i.e., TDL, ICP, EMU, PLN, and WST). However, TP concentrations across the least-
disturbed sites were significantly higher in tidal (TDL) compared to the inland subpopulations (i.e., ICP,
EMU, PLN, and WST), although TP did not differ significantly among the 4 inland reporting units. The
significant differences among these subpopulations warranted separate TN and TP stressor condition
thresholds for inland and tidal subpopulations (i.e., HYD_CLS, see Table 5-1 in Chapter 5:).
Thus, TN and TP thresholds for good stressor condition and poor stressor condition were developed for
inland and tidal subpopulations using the distribution of least-disturbed sites sampled for water
chemistry. After deleting outliers using a 1.5*IQR test (with IQR referring to "interquartile range"),
threshold values were calculated using the 75th and 95th percentiles of TN and TP concentrations among
least-disturbed sites sampled for water chemistry (Figure 13-2). Specifically:
• Good stressor condition thresholds were calculated as the 75th percentile of TN and TP
concentrations among least-disturbed sites sampled for water chemistry.
• Poor stressor condition thresholds were calculated as the 95th percentile of TN and TP
concentrations among least-disturbed sites sampled for water chemistry.
• Sites with TN and TP concentrations higher than the threshold for "good" and lower than the
threshold for "poor" are classified as fair stressor condition.
212
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Distribution of Least-Disturbed Sites
Sampled for Water Chemistry
(!)
D
Q.
CO
o
.c
a.
TO
-#-»
1°
c
a>
O)
o
+->
£
o
Percentiles:
^MrStressor Con_djtion
Fair Stressor Condition
Good Stress
or Condition
50th
25th
Figure 13-2. Good stressor condition and poor stressor condition threshold-setting using the 75th and 95th
percentiles of total nitrogen (TN) or total phosphorus (TP) concentrations among least-disturbed sites sampled for
water chemistry.
Threshold results for inland sites and tidal sites are shown in Table 13-7. Inland and tidal thresholds are
very similar for TN (approximately 1.2 mg/L for both inland and tidal sites) but very different for TP (98
|ag/L for inland sites and 174 |ag/L for tidal sites). In general, wetlands tend to have higher "natural"
background TN and TP concentrations compared to streams. For comparison, good stressor condition
thresholds for NARS streams in mountainous ecoregions (SAP, NAP, and WMT from the AG_EC09
subpopulation) are approximately 0.15-0.35 mg/L for TN and 15-20 |ag/L for TP, and approximately 0.6-
0.7 mg/L for TN and 50-90 |ag/L for TP in the Plains ecoregions (NPL, SPL, and TPL from the AG_EC09
subpopulation) (USEPA 2016d).
Table 13-7. Final total nitrogen (TN) and total phosphorus (TP) thresholds and relevant information for developing
those thresholds, including the number of least-disturbed sites with water chemistry on which threshold
percentiles are based (see Section 13.3 for details), the high outlier cut-off, and the number of outlier sites.
Total Nitrogen
Inland Sites
Total Nitrogen
Tidal Sites
Total Phosphorus
Inland Sites
Total Phosphorus
Tidal Sites
Number of Least-Disturbed
Sites with Water Chemistry
152
92
152
92
High Outlier Cut-off
3.073 mg/L
2.858 mg/L
240.5 |ig/L
424.0 |ig/L
Number of Outlier Sites
Removed from Analysis
14
10
9
7
Good (75th percentile)
Stressor Condition Threshold
< 1.26 mg/L
< 1.24 mg/L
< 98 |ig/L
< 174 |ig/L
Poor (95th percentile)
Stressor Condition Threshold
> 2.04 mg/L
> 2.18 mg/L
> 166 |ig/L
> 358 |ig/L
213
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
13.5 Applying TN and TP Stressor Condition Thresholds to the NWCA 2021
Water Chemistry Data
Water chemistry samples were collected and validated for NWCA 2021 using the procedures described in
Sections 13.1 and 13.2 for the same set of analytes listed in Table 13-1 (USEPA 2021a,b). The percentage
of 2021 sites with water chemistry samples was similar to 2016 with 63% of probability and handpicked
sites across both Visit 1 and Visit 2 yielding a water sample. The same TN and TP stressor condition
thresholds developed for the NWCA 2016 analysis were applied to calculate population estimates of
conditon and for change analysis between 2021 and earlier survey years.
13.6 Literature Cited
Dewitz J (2019) National Land Cover Database (NLCD) 2016 Products: U.S. Geological Survey data release,
https://doi.org/10.5066/P96H HBIE
Fennessy MS, Jacobs AD, Kentula ME (2007) An evaluation of rapid methods for assessing the ecological
condition of wetlands. Wetlands 27: 543-560
Herlihy AT, Kaufmann PR, Mitch ME (1990) Regional estimates of acid mine drainage impact on streams in
the Mid-Atlantic and Southeastern United States. Water, Air, Soil Pollution 50: 91-107
Herlihy AT, Stoddard JL, Burch Johnson C (1998) The relationship between stream chemistry and
watershed land cover data in the Mid-Atlantic Region, US. Water, Air, Soil Pollution 105: 377-386
Hornung RW, Reed LD (1990) Estimation of average concentration in the presence of nondetectable
values. Applied Occupational and Environmental Hygiene 5(1): 46-51
Jacobs AD (2007) Delaware Rapid Assessment Procedure Version 4.1. Delaware Department of Natural
Resources and Environmental Control, Dover, DE
Kaufmann PR, Levine P, Robison EG, Seeliger C, Peck DV (1999) Quantifying Physical Habitat in Wadable
Streams. EPA/620/R_99/003. US Environmental Protection Agency, Washington, DC
Kaufmann PR, Hughes RM, Van Sickle J, Whittier TR, Seeliger CW, Paulsen SG (2014) Lakeshore and littoral
physical habitat structure: a field survey method and its precision. Lake and Reservoir Management 30:
157-176
Paulsen SG, Mayio A, Peck DV, Stoddard JL, Tarquinio E, Holdsworth SM, Sickle JV, Yuan LL, Hawkins CP,
Herlihy AT, Kaufmann PR, Barbour MT, Larsen DP, Olsen AR (2008) Conditions of stream ecosystems in
the US: an overview of the first national assessment. Journal of the North American Benthological Society
27(4), 812-821
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
214
-------
Stoddard JL, Herlihy AT, Peck DV, Hughes RM, Whittier TR, Tarquinio E (2008) A process for creating
multimetric indices for large-scale aquatic surveys. Journal of North American Benthological Society 27:
878-891
USEPA (2016a) National Wetland Condition Assessment 2016: Field Operations Manual. EPA-843-R-15-
007. US Environmental Protection Agency, Washington, DC
USEPA (2016b) National Wetland Condition Assessment 2016: Laboratory Operations Manual. EPA-843-R-
15-009. US Environmental Protection Agency, Washington, DC
USEPA (2016c) National Wetland Condition Assessment: 2011 Technical Report. EPA-843-R-15-006. US
Environmental Protection Agency, Washington, DC
USEPA (2016d) National Rivers and Streams Assessment 2008-2009 Technical Report. EPA-841-R-16-008.
US Environmental Protection Agency, Washington, DC
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual (EPA-843-B-21-
002
USEPA (2021b) National Wetland Condition Assessment 2021: Laboratory Operations Manual (EPA-843-B-
21-003
Wardrop DH, Kentula ME, Jensen SF, Stevens Jr. DL, Hychka KC, Brooks RP (2007) Assessment of wetlands
in the Upper Juniata watershed in Pennsylvania, USA using the hydrogeomorphic approach. Wetlands 27:
432-445
Whigham DF, Jacobs AD, Weller DE, Jordan TE, Kentula ME, Jensen SF, Stevens Jr. DL (2007) Combining
HGM and EMAP procedures to assess wetlands at the watershed scale - Status of flats and non-tidal
riverine wetlands in the Nanticoke River Watershed, Delaware and Maryland (USA). Wetlands 27: 462-
478
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
215
-------
National Wetland Condition Assessment: 2021 Technical Support Document
Dec 2024
-------
Chapter 14: Microcystins
Microcystins are one group of naturally occurring toxins produced by various cyanobacteria (blue-green
algae) that are common to surface waters (Chorus and Bartram 1999). Microcystins have been detected
nationally in wetlands (USEPA 2016b, USEPA 2022a) and are considered to be the most commonly
occurring class of cyanobacteria toxins (cyanotoxins) (Chorus and Bartram 1999). Microcystin exposure
risk is typically elevated when an overabundance of cyanobacteria occurs in surface water causing a
cyanobacteria harmful algal bloom (cyanoHABs). There is concern that changes in weather patterns,
human population expansion, and associated behaviors are leading to perceived increases in occurrence
and severity of cyanoHABs (Paerl and Scott 2010). Three main exposure scenarios are of potential
concern regarding microcystins and wetlands: direct ecological impacts on plants and animals, human
consumption of exposed organisms, and direct human exposure through recreational contact.
Adverse ecological impacts due to microcystin exposure on plants and animals have been summarized in
several sources. Various adverse impacts of microcystins on cellular processes in a variety of aquatic and
terrestrial plants resulting in diminished plant growth and accumulation of microcystins have been
reported (Crush et al. 2008, Corbel et al. 2013, Romero-Oliva et al. 2014). Some macrophytes common to
certain types of wetlands have shown sensitivity to microcystins also. Microcystins have been shown to
inhibit the growth and oxygen production of some wetland macrophytes at concentrations of 1 |ag/L or
less (Rojo et al. 2013). Additionally, illness and mortality due to microcystin exposure has been reported
in wildlife, livestock, companion animals and all trophic levels of freshwater, brackish and marine aquatic
life. Animal illness and mortality has been reported in numerous cases including amphibians, cats, cattle,
chickens, deer, dogs, frogs, horses, muskrat, sheep, turkey, and waterfowl, but the true number of cases
remains unknown since many are not reported or observed (Chorus and Bartram 1999, Landsberg 2002,
Briand et al. 2003, Handeland and 0stensvik 2010, Vareli et al. 2013).
14.1 Data Collection and Analysis
Samples were collected for microcystin analysis from sites with sufficient surface water for sample
collection and shipped to analytical labs following procedures outlined in the NWCA Field Operations
Manual (USEPA 2021a). Samples were lysed by three sequential freeze/thaw cycles and filtered with 0.45
micron HVLP syringe filters (Loftin et al. 2008, Graham et al. 2010). Following the NWCA Laboratory
Operations Manual (USEPA 2021b), samples were analyzed by one of two methods depending on
whether practical salinity units (PSU) were < 3.5 PPT (part per thousand, Method 1) or > 3.5 PPT (Method
2). Samples were stored frozen prior to further extraction (Method 2) and analysis for microcystins by
enzyme-linked immunosorbent assay (Abraxis ADDA kit, Warminster, PA) at -20°C.
14.2 Application of EPA Recommended Criterion for Microcystins
Microcystins concentrations were evaluated against the EPA recommended recreational water quality
criterion and swimming advisory level of 8 ppb (US EPA 2019). Microcystins results identify the
percentage of wetland area at or below the criterion and above the criterion. The microcystins detection
217
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
results were determined using the MDL of 0.1 ppb which was consistent in both surveys. The detection
results presented in the public report and data dashboard represent the percentage of wetland area with
measured values greater than 0.1 ppb.
14.3 Literature Cited
Abraxis Bulletin (R110211) Microcystins in brackish water or seawater sample preparation.
http://www.abraxiskits.com/uploads/products/docfiles/385_MCT-
ADDA%20in%20Seawater%20Sample%20Prep%20%20Bulletin%20R041112.pdf: accessed on 2/20/2015
Briand JF, Jacquet S, Bernard C, Humbert JF (2003) Health hazards for terrestrial vertebrates from toxic
cyanobacteria in surface water ecosystems. Veterinary Research 34: 361-377
Corbel S, Mougin C, Bouaicha N (2013) Review: Cyanobacteria toxins: Mode of actions, fate in aquatic and
soil ecosystems, phytotoxicity and bioaccumulation in agricultural crops. Chemosphere 96: 1-15
Crush JR, Briggs LR, Sprosen JM, Nichols SN (2008) Effect of irrigation with lake water containing
microcystins on microcystin content and growth of ryegrass, clover, rape, and lettuce. Environmental
Toxicology 23: 246-252
Chorus I, Bartram J (Eds.) (1999) Toxic cyanobacteria in water: A guide to their public health
consequences, monitoring, and management. World Health Organization and E&FN Spon Press, London,
UK
Graham JL, Loftin KL, Meyer MT, Ziegler AC (2010) Cyanotoxin mixtures and taste-and-odor compounds in
cyanobacterial blooms from the Midwestern United States. Environmental Science and Technology 44:
7361-7368.
Handeland K, 0stensvik 0 (2010) Microcystin poisoning in roe deer (Capreolus capreolus). Toxicon 56:
1076-1078
Ibelings BW, Chorus I (2007) Accumulation of cyanobacterial toxins in freshwater "seafood" and its
consequences for public health: A review. Environmental Pollution 150: 177-192
Landsberg JH (2002) The effects of harmful algal blooms on aquatic organisms. Reviews in Fisheries
Science 10: 113-390
Loftin KA, Meyer MT, Rubio F, Kamp L, Humphries E, Wherea, E (2008) Comparison of two cell-lysis
procedures for recovery of microcystins in water samples from Silver Lake in Dover, Delaware with
microcystin producing cyanobacterial accumulations. US Geological Survey Open File Report 2008-1341
USEPA (2016b) National Wetland Condition Assessment 2011: A Collaborative Survey of the Nation's
Wetlands. EPA-843-R-15-005. US Environmental Protection Agency, Office of Water, Washington, DC
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual (EPA-843-B-21-
002
218
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
USEPA (2021b) National Wetland Condition Assessment 2021: Laboratory Operations Manual (EPA-843-B-
21-003
USEPA (2022a) National Wetland Condition Assessment: The Second Collaborative Survey of Wetlands in
the United States. EPA-841-R-23-001. US Environmental Protection Agency, Washington, DC
US EPA (2019) Recommended human health recreational ambient water quality criteria or swimming
advisories for microcystins and cylindrospermopsin. US EPA Office of Water. EPA 822-R-19-001.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
219
-------
National Wetland Condition Assessment: 2021 Technical Support Document
Dec 2024
-------
Chapter 15: Condition Extents, Change in Condition Extents, and
Relative and Attributable Risk
REPORTING (CHP 15)
and NWCA Public Reporting
1
DATA
PREPARATION
(CHP 4)
prepare & QA
data before use
SURVEY
DESIGN
(CHP 2)
Probability
sites (PROB)
HANDPICKED
SITES
(CHP 3)
handpicked
sites (HAND)
collect data
in the field
using protocol
from the Field
Operations
Manual
for each site, calculate metrics and indices,
including:
NONNATIVE PLANT
INDICATOR (NNPI) (CHP 10)
WATER CHEMISTRY (CHP 13)
MICROCYSTES (CHP 14)
HUMAN-MEDIATED
PHYSICAL ALTERATIONS (CHP 11)1
SOIL HEAVY METALS (CHP 12)H
PERCENT RELATIVE COVER OF
NONNATIVE PLANTS (SECT 6.6)
STRESSOR
THRESHOLDS
(CHP 10-14)
H
least disturbed
develop vegetation
multimetric indices
VEGETATION
ANALYSIS
OVERVIEW-,
DATA
ACQUISITION,
PREP (CHP 7)
~
PREREQUISITE
ANALYSES TO
VEGETATION
INDICATOR
DEVELOPMENT
(CHP 8)
~
site weights from
~~ probabiliy design
only probability sites
used for population
estimates:
SUBPOPULATIONS
(CHP 5)
Recap of Figure 1-1. Annotated analysis flow chart indicating the chapter number (abbreviated as "CHP") in which
details may be found.
The information provided in the previous chapters is intended to provide a solid understanding of how
the NWCA 2021 was designed, conducted, and analyzed. Up to this point in this document, details have
been provided regarding the:
• survey design (Chapter 2:),
• selection of handpicked sites (Chapter 3:),
• preparation of data (Chapter 4:),
• definition of subpopulations (Chapter 5:),
• establishment of the disturbance gradient (Chapter 6:),
• development of the Vegetation Multimetric Indices (VMMIs) (Chapter 7: through Chapter 9:),
• development of the Nonnative Plant Indicator (NNPI) (Chapter 10:), and
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
• development of physical and chemical indicators used for the disturbance gradient (Chapter 6:)
and stressor condition (Chapter 11: through Chapter 14:).
This chapter will describe how all the above components are used to calculate population estimates,
which include three different types of condition:
• wetland condition extent estimates based on the Vegetation Multimetric Indices (VMMIs)
(Section 15.1.1),
• Nonnative Plant Indicator (NNPI) condition extent estimates (Section 15.1.2), and
• stressor condition extent estimates based on physical and chemical indicators (Section 15.1.3).
Wetland condition, NNPI condition, and stressor condition extent estimates are calculated using spsurvey:
Spatial Survey Design and Analysis (Dumelle et al. 2023) and expressed as wetland area in acres or
percent of the resource; therefore, site weights from the probability design must be used to generate
population estimates along with the data from the probability sites sampled (n-sites = 933). The role of
population estimates and site weights in these calculations is discussed in Section 15.1. Additionally,
methods for calculating and reporting change in wetland condition and stressor condition extent
estimates between the NWCA 2011, NWCA 2016 and NWCA 2021 (referred to as "change analyses") are
discussed in Section 15.2. Ultimately, relative and attributable risk, discussed in detail in Section 15.3, are
used to calculate the relationship between:
• wetland condition and stressors, and
• NNPI condition and stressors.
Final results, including:
• wetland condition extent estimates,
• NNPI condition extent estimates,
• stressor condition extent estimates,
• change analyses, and
• relative and attributable risk
are presented in National Wetland Condition Assessment: The Third Collaborative Survey of Wetlands in
the United States (USEPA 2024a) and in the USEPA National Wetland Condition Assessment 2021 Data
Dashboard (2024b), primarily as bar graphs. This document provides guidance on how to interpret these
results.
15.1 Condition Extent Estimates
The survey design for the NWCA, discussed in Chapter 2: of this report, produces a spatially-balanced
sample using the USFWS National Wetland Inventory (NWI) (USFWS 2014). Each point (n-probability sites
= 933, see Table 6-1) has a known probability of being sampled (Stevens and Olsen 1999, Stevens and
Olsen 2000, Stevens and Olsen 2004, Olsen et al. 2019), and a sample weight is assigned to each
individual site as the inverse of the probability of that point being sampled. Sample weights are expressed
in units of acres.
The probability of a site being sampled, as discussed in Chapter 2:, Section 2.2.3, was stratified by state
with unequal probability of selection based on geographic regions and Wetland Groups (WETCLS_GRP)
see Table 5-1 in Chapter 5:) within each state. Site weights for the survey were adjusted to account for
222
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
additional sites (i.e., oversample points) that were evaluated when the primary sites were not sampled
(e.g., due to denial of access, being non-target). These site weights, designated by the red "W" enclosed
in a circle (i.e., ©) in the Overview of Analysis figure (Figure 1-1), are explicitly used in the calculation of
wetland condition extent estimates, NNPI condition extent estimates, and stressor condition extent
estimates, so results can be expressed as estimates of wetland area (i.e., numbers of acres or percent of
the entire resource) in a particular condition category (i.e., "good", "fair", "poor", and, for the NNPI only,
"very poor") for the Nation and any of the subpopulations in Table 5-1. In the following sections, the
methods by which estimates are calculated and reported are described for wetland condition extent
(Section 15.1.1), NNPI condition extent (Section ), and stressor condition extent (Section 15.1.3). It is
important to note that the NWCA was not designed to report on individual sites or states, but to report at
national and regional scales (see Chapter 2:).
15.1.1 Wetland Condition Extent Estimates
A Vegetation Multimetric Index (VMMI) summarizes several metrics describing different aspects of
observed vegetation that together can reflect wetland condition in relation to least-disturbed wetland
sites. For the NWCA 2021 analysis, four separate VMMIs, developed for previous NWCA analyses in 2016,
were used, one for each of four Wetland Groups: Estuarine Flerbaceous, Estuarine Woody, Inland
Flerbaceous, and Inland Woody.
Wetland condition extent estimates are based on the four Vegetation Multi metric Indices (VMMIs). Each
NWCA probability site is designated as in good, fair, or poor condition based on its VMMI value and
associated thresholds appropriate to the site (Chapter 7). Next, the site weights from the probability
design are summed across all sites in each condition category to estimate the wetland area in good, fair,
and poor condition for the NWCA target wetland population (see Chapter 2:, Section 2.2.5) nationally and
for the subpopulations reported in Table 5-1. The survey design allows calculation of confidence intervals
around these condition estimates.
Note that only Visit 1 (i.e., the Index Visit) data and only probability sites are used in the calculation of
extent. Flandpicked sites have a weight of zero. Using this method, wetland area in a particular wetland
condition category is estimated and reported in numbers of acres or by percent of the resource (Figure
15-1). The National Wetland Condition Assessment: The Third Collaborative Survey of Wetlands in the
United States (2024a) provides national results, whereas the USEPA National Wetland Condition
Assessment 2021 Data Dashboard (2024b) provides an interactive format for users to explore national
results and results for different subpopulations.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
223
-------
2021 National Extent Estimates for Wetland Condition
Based on the VMMI
Good
Fair
Poor
Not Assessed
45%
3-
20%
<1%
0%
20%
34%
40%
60%
80%
100%
Percent (%) of Target Wetland Population
Figure 15-1. The NWCA 2021 national extent estimates for wetland condition based on the Vegetation Multimetric
Indices (VMMIs). Wetland condition extent is presented for each condition category by percent of the resource
(i.e., percent of target wetland area for the Nation). Error bars represent 95% confidence intervals as calculated by
the R package spsurvey (Dumelle et al. 2023).
15.1.2 Normative Plant Indicator (NNPI) Condition Extent Estimates
Nonnative plant species are widely recognized as important biological indicators of lowered ecological
condition. They have numerous direct and indirect effects on native vegetation and other ecosystem
components, properties, and processes. The Nonnative Plant Indicator (NNPI) reflects wetland condition
in relation to stress from nonnative plants (Magee et al. 2019) by incorporating attributes of richness,
occurrence, and abundance for nonnative (alien and cryptogenic) plant species (see Chapter 10:).
NNPI condition extent estimates are based on the designation of each probability site as good, fair, poor,
or very poor condition based on NNPI. Site weights from the probability design are summed across all
sites in each condition category to estimate the wetland area in good, fair, poor, and very poor condition
for the NWCA target wetland population (see Chapter 2:, Section 2.2.5) nationally and for the
subpopulations reported in Table 5-1. The survey design allows calculation of confidence intervals around
these condition estimates.
Note that only Visit 1 (i.e., the Index Visit) data and only probability sites are used in the calculation of
extent. Handpicked sites have a weight of zero. Using this method, wetland area in a particular NNPI
condition category is estimated and reported in numbers of acres or by percent of the resource (Figure
15-2). The National Wetland Condition Assessment: The Third Collaborative Survey of Wetlands in the
United States (2024a) provides national results, whereas the USEPA National Wetland Condition
Assessment 2021 Data Dashboard (2024b) provides an interactive format for users to explore national
results and results for different subpopulations.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
224
-------
2021 National Extent Estimates for NNPI Condition
Good
— 48%
Fair
—|— 27%
Poor
- 13%
Very Poor
11%
Not Assessed
<1%
r i
i i i
1 1 1 1 1
0% 20% 40% 60% S0% 100%
Percent (%) of Target Wetland Population
Figure 15-2. The NWCA 2021 national extent estimates for Nonnative Plant Indicator (NNPI) condition. NNPI
condition extent is presented for each condition category by percent of the resource (i.e., percent of target
wetland area for the Nation). Error bars represent 95% confidence intervals as calculated by the R package
spsurvey (Dumelle et al. 2023).
15.1.3 Stressor Condition Extent Estimates
Indicators of stressor condition are used as descriptors of the potential impact of anthropogenic activities
on wetland condition. Although indicators of stressor condition do not necessarily imply causation of
ecological decline, they are often associated with impaired condition. For simplicity, they are sometimes
referred to using the shorthand term "stressors". Stressors are used to support analyses that provide four
types of information (i.e., results):
• Stressor Condition Extent-an estimate (by percent of the resource or relative ranking of
occurrence) of how spatially common a stressor is based on the population design;
• Relative Extent- an estimate of the areal percentage of the wetland population with poor
stressor condition for a particular indicator;
• Relative Risk -the probability (i.e., risk or likelihood) of having poor condition when the
stressor condition category is poor relative to when it is good; and,
• Attributable Risk - an estimate of the proportion of the population in poor condition that might
be reduced if the effects of a particular stressor were eliminated (Van Sickle and Paulsen 2008).
Ten indicators of stressor condition are reported for the NWCA 2021 (2024a,b):
• Vegetation Removal (VEGRMV),
• Vegetation Replacement (VEGRPL),
• Water Addition/Subtraction (WADSUB),
• Flow Obstruction (WOBSTR),
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
• Soil Hardening (SOHARD),
• Surface Modification (SOMODF), and
• Physical Alterations (PALT_SUM)
• Total Nitrogen in the water column (TN),
• Total Phosphorus in the water column (TP), and
• Microcystins (MICX).
Stressor condition categories are defined at each wetland site as "good", "fair", or "poor", except for
microcystins, which is defined as "at or below benchmark" or "above benchmark". These stressor
condition categories were assigned for multiple physical, chemical, and human-health indicators based on
specific thresholds, as described at the end of each of the individual chapters describing the indicators
(i.e., Chapter 10: through Chapter 14:). To calculate stressor condition extent estimates, site weights
were summed by stressor condition category and applied to the NWCA target wetland population
(Chapter 2:, Section 2.5) nationally and the subpopulations reported in Table 5-1 to estimate wetland
area in the good, fair, and poor stressor condition categories. The National Wetland Condition
Assessment: The Third Collaborative Survey of Wetlands in the United States (2024a) provides national
results, whereas the USEPA National Wetland Condition Assessment 2021 Data Dashboard (2024b)
provides an interactive format for users to explore national results and results for different
subpopulations.
Note that only Visit 1 (i.e., the Index Visit) data and only probability sites are used in the calculation of
extent. Handpicked sites have a weight of zero. Using this method, wetland area in a particular stressor
condition category is estimated and reported in numbers of acres or by percent of the resource (Figure
15-3). Population results for condition based on the 10 stressors are detailed in the National Wetland
Condition Assessment: The Third Collaborative Survey of the Nation's Wetlands (USEPA 2024a).
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
2021 National Extent Estimates for Stressor Condition
Good
—
4%
Good
— 79%
Far
+
17%
Fair
15%
Pock
- 7%
Poor
j*%
Not Assessed
J- 2%
WADSUB
Not Asm&mmI
h
WOBSTR
0* 2©* *0% 60* 50* 1«J% o* 20* 40* »* W* 100*
Good
— — 49^
Good
— — 74%
Fab
+-
Fa»
- 18%
Root
¦j- tr*
Poor
NotAa&essod
1»
SOHARD
Kbt Assessed
h
SOMODF
80% 140*
20* 40*
80* I OS*
Good
F*r
POOf
Not Assessed I™
40%
- 42%
PALT SUM
Ala Beta*
Criterion
Above Qiterfon
Not Assessed
MICX
o*
20* 40*
Good
— 29%
Fair
-7%
Poor
—j— 24%
Not Assessed
40%
TP
«* 2*%
40* W*
Figure 15-3, The NWCA 2021 national extent estimates for 10 indicators of stressor condition. Stressor condition
extent is presented for each condition category by percent of the resource (i.e., percent of target wetland area for
the Nation). Error bars represent 95% confidence intervals as calculated by the R package spsurvey (Dumelle et al.
2023). Stressor abbreviations are defined in Section 15.1.3.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
227
-------
15.2 Change and trend analysis
One of the objectives of the NWCA is to track changes in the condition of wetlands over time. For the
third cycle of the NWCA, change analyses were performed to determine the difference in the condition of
the wetland population between 2011-2021, and 2016-2021. Trend information was also calculated,
using a linear regression.
15.2.1 Data Preparation
Analyses focused on the change in condition from the 2021 survey and prior surveys (i.e., 2016 and
2011) and used data collected from all sites sampled in 2011 (n=967), 2016 (n=967), and 2021 (n=933).
15.2.2 Methods
Change analysis was conducted through the use of the spsurvey package in R (Dumelle et al. 2023).
Within the GRTS (Generalized Random Tessellation Stratified) survey design, change analysis can be
conducted on continuous or categorical variables (e.g., good, fair, and poor). The analysis measures the
difference between response variables of two survey time periods. For NWCA 2021, the categorical
response variables were used to compare changes between NWCA 2011 and 2021, and 2016 and 2021..
When using categorical variables, change is estimated by the difference in category estimates from the
two surveys. Category estimates were defined as the estimated proportion of values in each category
(i.e., good, fair, and poor (or very poor in the case of NNPI) categories). Change between the two years
was identified as statistically significant in the interactive data dashboard and web-report when the
resulting error bars around the change estimate did not cross zero. Statistical significance is provided as a
way to highlight results that may warrant additional exploration and analyses.
For some indicators and subpopulations, the change in the percentage of wetland area that is "not
assessed" can be relatively large and may change from survey to survey. Large changes in not assessed
may reflect changes in sampling or assessment success rather than actual changes in condition associated
with other categories such as good, fair and poor. Therefore, when the percent of not assessed increases
or decreases by more than 5 percentage points between survey cycles, EPA will not present these results
in the interactive dashboard to limit potentially erroneous interpretations of condition change.
Change estimates could not be made for some indicators and some survey cycles due to differences in
methodologies (e.g., physical alterations, soil heavy metals, water chemistry).
Change analysis is conducted between two points in time (n =2) and thus can only analyze differences
between two survey time periods. In other words, changes between 2011 and 2021 and between 2016
and 2021 do not necessarily indicate trend or pattern of change. Trends are likely to become clearer after
additional survey years (e.g., adding results for 2026).
15.3 Relative Extent, and Relative and Attributable Risk
The relationship between condition based on the VMMI or the NNPI and the condition based on
indicators of stress can be described by calculating relative extent, and relative and attributable risk.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
228
-------
15.3.1 Relative Extent
Relative extent shows the percent of the resource estimated to be in a given condition category for an
indicator. Here, the relative extent of poor stressor condition for a given indicator is calculated for
comparison to relative and attributable risk results (as shown in the left panels in Figure 15-4).
Relative Extent (Percentage of Wetland Area in Poor Condition)
Relative Risk
AttnhutaDle Risk
0% 20% 40% 80% 80% 100% 0 1 2 3 4 5
0% 20% 40% 00% S0% 100%
&—
Chemical water Quality Nitrogen | -f*
1 1 |
indicators
i
r
' ^
i
Alteration
i
Vegetation Removal | "4-
—¦
Flow Obstruction Q*
B-
water Addition/Subtraction [-
1 '
&
Soil Hardening [
f :
IZ3-
Surface Modification Q*
| +•—
S-
increased Risk
20% 40%
30% 100% 0
20% 40%
80% 100%
a.
z:
z
O.
>
o
2
Chemical water Quality Nitrogen
Indicators
Waler Quality Phosphorus
Physical Physical Alterations (Sum)
Alteration
Vegetation Removal
Flow Obstruct ton
Water Addrtioa'Subtraction
Soil Hardening
Surface Modification
b
~-
B
Q-
D-
H-
L
Increased Risk
Figure 15-4, The NVVCA 2021 relative extent of wetlands with stressors in poor condition, and the relative risk and
attributable risk of poor a) VMMI condition or b) NNPI condition when stressor condition is poor as calculated by
the R package spsurvey (Dumelle et al. 2023). For relative risk, values below the dashed line (i.e., a relative risk < 1)
signifies that there is no association between the stressor and VMMI or NNPI condition.
15.3.2 Relative Risk
Relative risk is the probability (i.e., risk or likelihood) of having poor wetland condition based on the
VMMI, or poor/very poor NNPI condition, when the stressor condition category is poor relative to when
the stressor condition category is good. Relative risk analysis was derived from medical literature, where
it is used commonly to describe, for example, the risk of having a heart attack based on cholesterol levels.
The fact that relative risk is used so commonly to report human health risks is an advantage because, as a
result, relative risk is an understandable concept to the general public. Applied to the NWCA, a relative
risk analysis can be used to evaluate the relative effect of a stressor on wetland condition based on the
VMMI or NNPI condition. Relative risk analyses are standard for reporting results in NARS assessments
(e.g., USEPA 2006; USEPA 2009), and examples can be found for lake and stream NARS assessments in the
literature (e.g., Van Sickle et al. 2006; Van Sickle et al. 2008; Van Sickle 2013).
229
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
15.3.2.1 Example Calculation of Relative Risk
Risk is calculated using contingency tables and expressed as a probability, which is unitless. Consider the
example two-by-two contingency table10 presented as Table 15-1, which relates stream condition
indicated by Fish Index of Biotic Integrity (IBI) and stress indicated by total nitrogen (TN). The probabilities
in the contingency table are calculated from weighted analysis of the data and reflect the proportion of
the resource, stream length in the case of Table 15-1, which is in each of the four cells of the table. For
wetland analysis, the resource is areal and the probabilities would reflect the proportion of wetland area
in the population in each of the cells.
Table 15-1. Example contingency table for relative risk that reports the proportion of stream length associated
with good and poor condition (as indicated by Fish Index of Biotic Integrity, IBI) and good and poor stressor
condition (as indicated by stream water total nitrogen concentration, TN). Results are hypothetical.
STRESS LEVEL
O
u
TN: Good TN:Poor
Fish IBI: Good
Fish IBI: Poor
0.598 0.275
0.070 0.056
Total
0.668 0.331
Using the hypothetical example data provided in Table 15-1, the risk of a stream having poorf\s\\
condition when the TN stressor condition is poor is calculated as:
0.056
0.331
= 0.169
The risk of a stream having poor condition when the TN stressor condition is good is calculated in the
same manner:
0.070
0.668
= 0.105
By comparing these two results, it is apparent that the risk of a stream having poor condition when the
TN stressor condition is poor (0.169) is greater than when the TN stressor condition is good (0.105). The
relative risk (RR) can then be simply calculated as the ratio of these two probabilities (Pr):
RR =
Pr (Poor condition given Poor stressor condition) 0.169
Pr (Poor condition given Good stressor condition) 0.105
= 1.61
Therefore, in this example, we can conclude that the risk of poor condition is 1.61 times greater in
streams with poor TN stressor condition than in streams with good TN stressor condition.
These calculations are repeated for each appropriate indicator of stress so relative risk can be reported
for each of them. If the stressor has no effect on condition, the relative risk is 1. Confidence intervals are
also used in reporting to express uncertainty in the estimate of relative risk (see Van Sickle et al. 2006).
10 The numbers used in this example are hypothetical and were not measured as part of any USEPA NARS
assessment.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
15.3.2.2 Considerations When Calculating and Interpreting Relative Risk
It is important to understand that contingency tables are created using a categorical, two-by-two matrix;
therefore, only two condition categories can be used. There are multiple methods by which condition
categories can be used for contingency tables.
For wetland condition categories based on the VMMI / condition categories for stressor indicators, three
methods of calculating contingency tables may be considered:
• Good vs. Poor / Good vs. Poor,
• Good vs. Not-Good / Good vs. Not-Good, or
• Not-Poor vs. Poor / Not-Poor vs. Poor*
where, "Not-Good" combines fair and poor condition categories, and "Not-Poor" combines good and fair
condition categories.
For NNPI condition categories/ condition categories for stressor indicators, five methods of calculating
contingency tables may be considered:
• Good vs. Very Poor/ Good vs. Poor
• Good vs. Poor + Very Poor / Good vs. Poor
• Good vs. Not-Good / Good vs. Not-Good
• Good + Fair vs. Poor + Very Poor / Not-Poor vs. Poor*
• Not-Very Poor vs. Very Poor / Not-Poor vs Poor
where, "Not-Good" combines fair, poor, and very poor condition categories, and "Not-Very Poor"
combines good, fair, and poor condition categories.
In the first bulleted method, "Good vs. Poor / Good vs. Poor", for example, data associated with the fair
condition categories are excluded from the analysis. Therefore, the results of the associated calculation of
relative risk are affected by which one of the above combinations is used to make the contingency tables,
and it is crucial that the objectives of the analysis are carefully considered to help guide this decision.
A second consideration is that relative risk does not model joint effects of correlated stressors. In other
words, each stressor is modeled individually, when in reality, stressors may interact with one another
potentially increasing or decreasing impact on condition. This is an important consideration when
interpreting the results associated with relative risk.
The two bold, asterisked (*) methods (one for each the VMMI and the NNPI condition categories) indicate
the method used for the NWCA analysis.
15.3.2.3 Application of Relative Risk to the NWCA
For each site sampled as part of the NWCA:
• Wetland condition is assigned as good, fair, or poor using the Vegetation Multimetric Index
(VMMI) thresholds as described in Chapter 9:;
• Nonnative Plant Indicator (NNPI) condition is assigned as good, fair, poor, or very poor, using
exceedance values as described in Chapter 10:; and
• Stressor conditions of physical and chemical indicators are assigned as good, fair, or poor using
thresholds as described in Chapter 11: through Chapter 14:.
231
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
For each indicator of stressor condition, a contingency table was created, comparing:
• the Not-Poor VMMI condition category (i.e., a combination of good and fair wetland conditions)
to Poor condition category, and Not-Poor stressor condition category (i.e., a combination of good
and fair stressor conditions) to Poor stressor condition; and
• the combination of Good and Fair NNPI condition categories to Poor and Very Poor NNPI
condition categories, and Not-Poor stressor condition category (i.e., a combination of good and
fair stressor conditions) to Poor stressor condition.
These decisions for the contingency tables were made because the objective of reporting relative risk in
the NWCA is to indicate which stressors policy makers and managers may want to prioritize for
management efforts to improve poor wetland condition. After creating contingency tables, relative risk
for each indicator of stress was calculated. Figure 15-4 provides the relative risk reported for the NWCA
2021; with stressor extent, relative risk provides an overall picture of the relative importance of individual
stressors on condition.
A relative risk value of 1.0 indicates that there is no association between the stressor and the VMMI or
NNPI, while values greater than 1.0 suggest greater relative risk. For example, if 30% of the population is
in poor condition based on the VMMI or NNPI, but the population is equally divided among sites with
Poor and Not-Poor stressor conditions (15% in each), then the RR = 0.15/0.15 = 1, and there is no
association between condition and the stressor. Conversely, if the 30% in poor condition was observed as
25% in sites with Poor stressor condition and 5% in sites with Not-Poor stressor condition, then the RR =
25/5 = 5.0. The higher the relative risk value for a given stressor, the greater the risk of poor wetland
condition. A relative risk of 5 indicates that we are five times more likely to see a wetland in poor
condition when the stressor is poor than when it is not poor (Herlihy et al. 2019).
15.3.3 Attributable Risk
Attributable risk provides an estimate of the proportion of the resource population (i.e., extent) in poor
condition that might be reduced if the effects of a particular stressor were eliminated. Attributable risk
(AR) combines relative stressor extent with relative risk into a single index using the following formula
(see Van Sickle et al. 2008 for details):
Pr(Extent with Poor Stressor Condition) * (RR — 1)
1 + Pr (Extent with Poor Stressor Condition) * (RR — 1)
where RR is relative risk and Pr is probability.
Similar to the consideration presented in Section 15.3.1.2, it is critical to define relative extent (i.e.,
percent of the resource) and relative risk in the same way. Therefore, for the NWCA data, the same
categories were used for calculating attributable risk as relative risk (e.g., Not-Poor was compared to
Poor) for VMMI and NNPI condition categories vs. stressor condition categories).
The ranking of stressors according to attributable risk (e.g., Figure 15-4) represents their relative
magnitude or importance relative to decreased ecological condition and can be used by policy makers
and managers to inform prioritization of actions for specific stressors, geographic area, and/or wetland
type.
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
232
-------
15.3.3.1 Considerations When Interpreting Attributable Risk
To appropriately interpret attributable risk, it is important to understand that attributable risk is
associated with the following three major assumptions:
• Causality, or that the stressor causes an increased probability of poor condition;
• Reversibility, or that if the stressor is eliminated, causal effects will also be eliminated and
damage is reversible; and,
• Independence, or that stressors are independent of each other, so that individual stressor effects
can be estimated in isolation from other stressors.
These assumptions should be kept in mind when applying attributable risk results to management
decisions. Attributable risk provides much needed insight into how to prioritize management for the
improvement of our Nation's aquatic ecosystems - wetlands, in the case of the NWCA. While the results
of attributable risk estimates are presented as percent area in poor condition that could be reduced if the
effects of a particular stressor were eliminated, these estimates are meant to serve as general guidance
as to what stressors are affecting condition and to what degree (relative to the other stressors evaluated).
15.4 Where to Find the Summary of NWCA Results
All of the methods presented in this document are the scientific basis for what is reported in National
Wetland Condition Assessment: The Third Collaborative Survey of Wetlands in the United States (USEPA
2024a) The report provides an overview of the important results from the NWCA 2021. The presentation
of results in that document is geared toward the lay public, environmental managers, and government
decision makers.
15.5 Literature Cited
Dahl TE (2006) Status and trends of wetlands in the conterminous United States 1998 to 2004. US
Department of Interior, Fish and Wildlife Service, Washington, DC
Dahl TE (2011) Status and trends of wetlands in the conterminous United States 2004 to 2009. US
Department of Interior, Fish and Wildlife Service, Washington, DC
Dahl TE, Bergeson MT (2009) Technical procedures for conducting status and trends of the Nation's
wetlands. US Fish and Wildlife Service, Division of Habitat and Resource Conservation, Washington, DC
Dumelle, M., T. Kincaid, A.R. Olsen, and M. Weber. 2023. Spsurvey: Spatial Sampling Desing and Analysis
in R. Journal of Statistical Software, 105(3), 1-29. https://doi.orR/10.18637/iss.vl05.i03
Herlihy AT, Paulsen SG, Kentula ME, Magee TK, Nahlik AM, Lomnicky GA (2019) Assessing the relative and
attributable risk of stressors to wetland condition across the conterminous United States. Environmental
Monitoring and Assessment 191 (SI): 320, doi: 10.1007/sl0661-019-7313-7
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
233
-------
Magee TK, Blocksom KA, Herlihy AT, &. Nahlik AM (2019) Characterizing nonnative plants in wetlands
across the conterminous United States. Environmental Monitoring and Assessment 191 (SI): 344, doi:
10.1007/s 10661-019-7317-3. https://link.springer.com/article/10.1007/sl0661-019-7317-3
Olsen AR, Peck DV (2008) Survey design and extent estimates for the Wadeable Streams Assessment.
Journal of the North American Benthological Society 27: 822-836
Olsen AR, Kincaid TM, Kentula ME, Weber MH (2019) Survey design to assess condition of wetlands in the
United States. Environmental Monitoring and Assessment 191 (SI): 268, doi: 10.1007/sl0661-019-7322-
6.
Stevens DL, Jr, Jensen SF (2007) Sampling design, implementation, and analysis for wetland assessment.
Wetlands 27: 515-523
Stevens DL, Jr, Olsen AR (1999) Spatially restricted surveys over time for aquatic resources. Journal of
Agricultural, Biological, and Environmental Statistics 4: 415-428
Stevens DL, Jr, Olsen AR (2000) Spatially restricted random sampling designs for design-based and model
based estimation. Pages 609-616 in Accuracy 2000: Proceedings of the 4th International Symposium on
Spatial Accuracy Assessment in Natural Resources and Environmental Sciences. Delft University Press, The
Netherlands
Stevens DL, Jr, Olsen AR (2004) Spatially-balanced sampling of natural resources. Journal of American
Statistical Association 99: 262-278
Van Sickle J, Stoddard JL, Paulsen SG, Olsen AR (2006) Using relative risk to compare the effects of aquatic
stressors at a regional scale. Environmental Management 38: 1020-1030
Van Sickle J, Paulsen SG (2008) Assessing the attributable risks, relative risks, and regional extents of
aquatic stressors. Journal of the North American Benthological Society 27: 920-931
Van Sickle J (2013) Estimating the risks of multiple, covarying stressors in the National Lakes Assessment.
Freshwater Science 32: 204-216
USEPA (2006) Wadeable Streams Assessment: A Collaborative Survey of the Nation's Streams. US
Environmental Protection Agency, Office of Water and Office of Research and Development, Washington,
DC
USEPA (2009) National Lakes Assessment: A Collaborative Survey of the Nation's Lakes. US Environmental
Protection Agency, Office of Water and Office of Research and Development, Washington, DC
USEPA (2016a) National Wetland Condition Assessment: 2011 Technical Report. EPA-843-R-15-006. US
Environmental Protection Agency, Washington, DC
USEPA (2016b) National Wetland Condition Assessment 2011: A Collaborative Survey of the Nation's
Wetlands. EPA-843-R-15-005. US Environmental Protection Agency, Office of Water, Washington, DC
234
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
USEPA (2021a) National Wetland Condition Assessment 2021: Field Operations Manual (EPA-843-B-21-
002
USEPA (2021b) National Wetland Condition Assessment 2021: Laboratory Operations Manual (EPA-843-B-
21-003
USEPA (2024a) National Wetland Condition Assessment: The Third Collaborative Survey of Wetlands in
the United States. EPA-843-R-24-001. US Environmental Protection Agency, Washington, DC
USEPA (2024b) National Wetland Condition Assessment2021 Data Dashboard. US Environmental
Protection Agency, Washington, DC
USFWS (2014) National Wetlands Inventory, US Department of the Interior, Fish and Wildlife Service,
Washington, D.C. https://www.fws.Rov/wetlands/Data/Metadata.html
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
235
-------
Glossary
Abiotic disturbance class- disturbance class assignments to sites based on only the physical and chemical
disturbance gradient screens (and not the biological disturbance gradient screen); the parameter name
for these abiotic disturbance class assignments is REF_NWCA_ABIOTIC
Assessment Area (AA)-Xhe 0.5 ha area that represents the location defined by the coordinates
generated by the NWCA sample draw, and in which most of the data collection for the NWCA occurs
Attributable Risk-an estimate of the proportion of the population in poor condition that might be
reduced if the effects of a particular stressor were eliminated11
Buffer- the area (representing a prescribed measurement area) surrounding the Assessment Area
Coefficients of Conservatism - (C-values, also called CCs) describe the tendency of individual plant species
to occur in disturbed versus near pristine conditions; C-values for individual species are state or regionally
specific and scaled from 0 to 10
Condition Category- describes the ecological condition of wetlands based on a biological indicator, a
Vegetation Multimetric Index (VMMI); classes include "Good", "Fair", or "Poor"
Condition Extent-estimates of the wetland area in good, fair, and poor condition categories
Contingency table- a two-by-two table that relates condition and stress used to calculate relative risk;
results of the contingency table are expressed as probabilities
Disturbance Class-classes reflecting the gradient of anthropogenic disturbance across all sampled
wetland sites, and used for the Vegetation Multimetric Index (VMMI) development and to set thresholds
for indicators of stress and condition
• Least Disturbed- a disturbance class describing sites that represent the best available physical,
chemical, and biological conditions in the current state of the landscape4; used as "reference" for
the NWCA Survey
• Most Disturbed - a disturbance class describing sites defined as most disturbed relative to "least
disturbed"; typically representing 20-30% of sites in an NWCA Reporting Group
• Intermediately Disturbed- a disturbance class used to describe sites that fall between "least
disturbed" and "most disturbed"
Disturbance gradient- a continuous gradient of anthropogenic disturbance, divided into three
disturbance classes to which wetland sites are assigned
11 Van Sickle J, Paulsen SG (2008) Assessing the attributable risks, relative risks, and regional extents of aquatic
stressors. Journal of the North American Benthological Society 27: 920-931
236
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Duration- longevity for plants, described by annual, biennial, and perennial life cycles or combinations
thereof (see Table 7-2 for details)
Exceedance value-for the NNPI, the exceedance of a threshold value for a particular condition category
for any one of the three metrics, resulting in the assignment of the metric condition to next lower (better
to worse) category; the NNPI condition for a site is based on the lowest observed condition category
among the metrics
Final disturbance class- disturbance class assignments to sites based on physical, chemical, and for least-
disturbed sites, biological disturbance gradient screens; the parameter name for these final disturbance
class assignments is REF_NWCA
Growth habit- Primary growth-habit types for the plant taxa (see Table 7-1 for details)
Handpickedsites- sampled sites suggested by states, tribes, and other partners based on the expectation
that they are minimally disturbed and can be used as least-disturbed (or "reference") sites
Indicator- a metric or index that reflects anthropogenic (human-mediated) disturbance to wetland
condition, vegetation condition, or stressor condition
Index- a combination of metrics used to generate a single score to describe a particular property
(disturbance, stressor condition, or wetland condition in the case of the NWCA) for a site
index period- the temporal range when sites were sampled for the 2011 NWCA; the peak growing
season (April through September, depending on state) when most vegetation is in flower or fruit
index Visit- the sampling event used when conducting analyses on the set of unique sites sampled
inference population -final wetland area represented by sampled probability sites; ultimately used by the
NWCA for reporting condition and stressor extent
Metric- an individual measurement or combinations of data types to describe a particular property (e.g.,
soil phosphorus concentration, species richness, species cover by growth form, etc.) for a site
Native Status- state level designations of plant taxa nativity for the NWCA, designations include:
• Native-plant taxa indigenous to specific states in the conterminous US
• Alien-combination of introduced and adventive taxa
• Introduced- plant taxa indigenous outside of, and not native in, the conterminous US
• Adventive- plant taxa native to some areas or states of the conterminous US, but introduced in
the location of occurrence
• Cryptogenic- plant taxa that includes both Native and Alien genotypes, varieties, or subspecies
• Undetermined-taxa identified at level of growth form, most families, or genera with both native
and alien species
Nonnativeplants-for the NWCA, includes both alien and cryptogenic taxa
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
237
-------
Oversample sites- a panel of additional sites selected by the survey design to provide replacements for
any sites that were either not part of the target population or could not be sampled
Parameter Names- specific code names (usually written in all caps) used to reference data in the official
NARS databases and in the NWCA raw datasets
Points-site coordinates selected by the survey design
Population- see the definition for "Target Population" in this Glossary
Population estimates- estimates of characteristics of the target or inference population of wetlands in
the conterminous US (or smaller reporting groups), usually described in acres or percent total area
Probability sites- sites defined by the NWCA sample draw (i.e., NWCA design sites) and some state
intensifications using the same design as NWCA
Reference-analogous to "least disturbed". Sites that represent least disturbed ecological condition12 and
the associated functional capacity typical of a given wetland type in a particular landscape setting (e.g.,
ecoregion, watershed)
Relative Extent- shows the percent of the resource estimated to be in a given condition category for an
indicator
Relative Risk (RR) - the probability (i.e., risk or likelihood) of having poor condition when the magnitude of
a stressor is high (i.e., poor stressor condition) relative to when the magnitude of a stressor is low (i.e.,
good stressor condition)
Resampiesites- probability sites that were originally sampled in the field in the previous NWCA survey
and selected to be sampled again in the current survey
Resource-the population of the aquatic resource (i.e., wetlands) evaluated in the NWCA
Revisit sites - a site sampled twice within the same year to assess within-season-variability in the
collected data
Sample frame-the geographic data layers that identify locations and boundaries of all wetlands that
meet the definition of the target population
Screen-the method for determining threshold (a.k.a., "cutoff" or "exceedance") values for assigning
disturbance class or condition category
Stressor Condition Extent-an estimate (by percent of the resource or relative ranking of occurrence, or
stressor-level class) of how spatially common a stressor is based on the population design
12 Stoddard JL, Larsen DP, Hawkins CP, Johnson PK, Norris RH (2006) Setting expectations for the ecological
condition of streams: the concept of reference condition. Ecological Applications 16:1267-1276
238
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
-------
Stressor Condition Category- describes the stress to wetlands associated with physical and chemical
indicators as "Good", "Fair", or "Poor"
Subpopuiations- individual units within a subpopulation group
Subpopuiation Group-the descriptive name for a parameter name and set of individual subpopuiations
Survey design - the methods by which sites are selected for the survey; in the case of the NWCA, a
Generalized Random Tessellation Stratified (GRTS) survey design is used,which provides spatially-
distributed samples that are more likely to be representative of the population than other common
spatial survey designs
Target population - also called "the population", all wetland area included in the NWCA Wetland Types
and used in the survey design; defined as all tidal and nontidal wetted areas with rooted vegetation and,
when present, shallow open water less than 1 meter in depth, and not currently in crop production,
across the conterminous US
Taxon-iocationpair- A particular plant taxon occurring at a particular location:
• X-regionpairs- where Xcan be any particular taxon, species, or name (e.g., one of several
potential taxonomic names) that occurs or was observed in a given region
• X-statepairs- where Xcan be any particular taxon, species, or name{e.g., one of several
potential taxonomic names) that occurs or was observed in a given state
• X-sitepairs- where Xcan be any particular taxon, species, or name[e.g., one of several potential
taxonomic names) that occurs or was observed in a given site
• X-piotpairs- where Xcan be any particular taxon, species, or name [e.g., one of several potential
taxonomic names) that occurs or was observed in a given plot
Thresholds- similar to "exceedance values" and analogous to "benchmarks", thresholds are specific
values used to delineate boundaries to assign sites to specific disturbance classes or condition categories
Unique sites- each unique site occupies the same coordinates but may have up to four sampling visits
(revisit sites (Visit 1 and Visit 2 in the same year) and resample sites (sites sampled in both 2011 and again
in 2016)
Wetland Indicator Status (WIS)- hydrophytic status for plants designated as one of seven WIS Categories
(see Table 7-3 for details)
Dec 2024
National Wetland Condition Assessment: 2021 Technical Support Document
239
------- |