xvEPA
United States Office of Research and EPA/620/R-97/002
Environmental Protection Development February 1997
Agency Washington DC 20460
Pilot Test of
Wetland Condition
Indicators in the Prairie
Pothole Region of the
United States
Missouri Drift Red River
Coteau Plain Valley
Environmental Monitoring and
Assessment Program
-------
PILOT TEST OF
WETLAND CONDITION INDICATORS
IN THE PRAIRIE POTHOLE REGION OF THE
UNITED STATES
EDITORS:
Spencer A. Peterson
U.S. Environmental Protection Agency
National Health and Environmental Effects Research Laboratory
Western Ecology Division
200 SW 35th Street
Corvallis, OR 97333
Lynn Carpenter
Agora Publishing Company
1217 St. Paul Street
Baltimore, MD 21202
Glenn Guntenspergen and Lewis M. Cowardin
U.S. Geological Survey
Biological Resources Division
Northern Prairie Science Center
Jamestown, ND 58401-7317
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Sectbn
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
9.1
9.2
10.0
11.0
Appendix 9-2-2
AUTHORS AND CONTRIBUTORS:
Lewis M. Cowardin1 and Spencer A. Peterson2
Lewis M. Cowardin1
David P. Fellows1 and Thomas K. Buhl1
Lewis M. Cowardin1 and Thomas Sklebar1
Ned H. Euliss1 and David M. Mushet1
Harold A. Kantrud1
John A. Freeland3 and Jim L. Richardson3
Diane L. Larson1
Ned H. Euliss1 and David M. Mushet1
Diane L. Larson1
Lewis M. Cowardin1
Lewis M. Cowardin1
Steve Dominguez2, Anne Fail-brother2 and Diane L. Larson1
1 U.S. Geological Survey, Biological Resources Division, Jamestown, ND 58401-7317
2 U.S. Environmental Protection Agency, National Health and Environmental Effects Research
Laboratory, Western Ecology Division, 200 SW 35th St., Corvallis, OR 97333
3 North Dakota State University, Department of Soil Science, Fargo, ND 58105-5638
-------
EPA/620/R-97/002
February 1997
PILOT TEST OF
WETLAND CONDITION INDICATORS
IN THE PRAIRIE POTHOLE REGION OF THE
UNITED STATES
Prepared by the
U.S. Geological Survey
Biological Resources Division
Northern Prairie Science Center
Jamestown, North Dakota
for the
U.S. Environmental Protection Agency
Office of Research and Development
National Health and Environmental Effects Research Laboratory
Western Ecology Division
Corvallis, Oregon
Spencer A. Peterson, Lynn Carpenter,
Glenn Guntenspergen and Lewis M. Cowardin, editors
December 1996
W9 Printed on Recycled Paper
-------
ACKNOWLEDGEMENTS
This report is a combined effort of the Biological Resources Division (BRD), USGS,
Northern Prairie Science Center (NPSC) at Jamestown, North Dakota, and the
Environmental Protection Agency (EPA), Office of Research and Development (ORD),
National Health and Environmental Effects Research Laboratory (NHEERL), Western
Ecology Division (WED) at Corvallis, Oregon. The work reported here was conducted
by staff of the NPSC, was initially assembled and edited by Lewis Cowardin and
Glenn Guntenspergen of the NPSC and is part of the EPA's Environmental Monitoring
and Assessment Program (EMAP) whose mission it is to develop, evaluate and
demonstrate the utility of environmental indicators to assess condition of various
environmental resource classes (Wetlands in this report). Participating persons and
organizations included
• NBS staff and seasonal hires in Jamestown, ND
• EMAP-Surface Waters and associated laboratory staff in Corvallis, OR,
including EPA and on-site contractor personnel (Dynamac International,
Inc., OAO, Inc.) and guest workers from Oregon State University
• US Fish and Wildlife Service, National Wetlands Inventory staff and
contractors at St. Petersburg, FL
• personnel on cooperative agreements with North Dakota State
University at Fargo
• members of the project peer review panel.
We especially appreciate the Northern Prairie Research Advisory Committee for their
suggestions on enhancement of the project proposal. The following people provided
official technical review of this report: Naomi Detenbeck, Louisa Squires and Daniel
Hubbard. Marge Hails did final word processing and formatting of the report.
Financial support for this project was shared by the
• US EPA, NHEERL, WED, Corvallis, OR under Interagency Agreement
number DW14935541-01.
• USGS, BRD, NPSC, Jamestown, ND (formerly U.S. Fish and Wildlife
Service under USFWS reference number 14-48-0009-92-1929.
All research reported here was conducted under the guise of a Quality Assurance
Research Plan prepared by NPSC and approved by EPA. This document has been
subjected to EPA's peer and administrative review and approved for publication as an
EPA document. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
-------
CONTENTS
SUMMARY xv
1.0 INTRODUCTION 1
1.1 OBJECTIVES OF EMAP PROGRAM 1
1.2 PRAIRIE POTHOLE REGION 2
1.3 INDICATORS OF CONDITION 3
1.4 RATIONALE FOR THE PILOT STUDY 5
1.5 OBJECTIVES OF PILOT STUDY 8
1.6 ORGANIZATION OF THE REPORT 9
2.0 DESIGN METHODS 11
2.1 REGIONALIZATION 11
2.2 STUDY PLOTS 11
2.3 METHOD OF SELECTING PLOTS TO REPRESENT EXTREMES IN CONDITION
(CROPLAND/UPLAND RATIO) 12
2.4 SELECTION OF SAMPLE WETLAND BASINS WITHIN PLOTS 15
2.5 SELECTION OF REPLACEMENT WETLAND BASINS AND PLOTS 17
2.6 REASSIGNMENT OF WETLAND BASIN CONDITION 17
3.0 ACCESS TO PRIVATE LAND AND LOGISTICS 21
3.1 RESEARCH ACCESS TO PRIVATELY OWNED WETLAND BASINS IN THE
PRAIRIE POTHOLE REGION OF THE UNITED STATES 21
3.2 IMPLICATIONS FOR FUTURE PROJECTS 21
4.0 TESTS OF SELECTED LANDSCAPE INDICATORS OF WETLAND CONDITION 23
4.1 OBJECTIVES 24
4.2 METHODS 25
4.2.1 Base Mapping of 10.4-km2 Plots 25
4.2.2 Aerial Video 28
4.2.3 Aerial Photographs 32
4.2.4 Analyses 35
4.2.5 Wetland Drainage Basins 37
4.2.6 Duck Populations and Production 38
4.3 RESULTS 39
4.3.1 Wetland Abundance and Distribution on Sample Areas 39
4.3.2 Drainage as An Indicator of Wetland Landscape Condition 42
4.3.3 Seasonal and Annual Change in Ponds 43
4.3.4 Upland Characteristics of Study Sites 45
4.3.5 Duck Populations and Production 50
4.4 EVALUATION AND RECOMMENDATIONS 50
4.4.1 Drainage as An Indicator of Condition 52
4.4.2 Area of Exposed Soil 56
4.4.3 Index to Wetland Change 56
4.4.4 Estimates of Duck Production 57
IV
-------
5.0 IMPACT OF AGRICULTURAL PRACTICES ON WETLAND MACROINVERTEBRATES,
SILTATION RATES, AND WATER-LEVEL FLUCTUATIONS 59
5.1 INTRODUCTION 59
5.2 OBJECTIVES 59
5.3 METHODS 60
5.3.1 Objective 1 60
5.3.2 Objective 2 62
5.3.3 Objective 3 63
5.4 RESULTS 63
5.4.1 Objective 1 63
5.4.2 Objective 2 67
5.4.3 Objective 3 67
5.5 EVALUATION 67
5.5.1 Objective 1 67
5.5.2 Objective 2 68
5.5.3 Objective 3 69
6.0 PLANTS AS INDICATORS OF WETLAND CONDITION IN THE PRAIRIE
POTHOLE REGION 71
6.1 INTRODUCTION 71
6.2 INDICATORS TESTED 73
6.3 METHODS 74
6.3.1 Design 74
6.3.2 Field Methods 77
6.3.2.1 Watershed and Wetland Classification 77
6.3.2.2 Plant Abundance and Species Richness 78
6.3.2.3 Standing Dead Vegetation and Litter Depth 81
6.3.3 Analysis 82
6.4 RESULTS 83
6.4.1 Watershed and Basin Classification 83
6.4.2 Community Characteristics 86
6.4.2.1 Distribution of Communities Among Wetland Zones and Phases . . 86
6.4.2.2 Land Use of Wetland Zones 91
6.4.2.3 Land Use of Watersheds 91
6.4.2.4 Physical Measurements in Communities 95
6.4.2.5 Botanical Measurements of Communities 98
6.5 EVALUATION 102
6.6 RECOMMENDATIONS FOR FUTURE EMAP STUDIES 110
7.0 SOILS AND SEDIMENTS AS INDICATORS OF AGRICULTURAL IMPACTS ON
NORTHERN PRAIRIE WETLANDS 119
7.1 INTRODUCTION 119
7.2 OBJECTIVES 120
7.3 METHODS 120
7.3.1 Quality and Quantity of Sediments 120
7.3.2 Long-term Sedimentation (Cottonwood Lake Study Area) 121
7.3.3 Key Soil Constituents 124
7.3.3.1 Cottonwood Lake Study Area 124
7.3.3.2 10.4-km2 Sample Plots 127
7.3.4 Soil Oxidation-Reduction 128
-------
7.4 RESULTS 129
7.4.1 Seasonal Sedimentation 129
7.4.2 Long-term Sedimentation 129
7.4.3 CWLSA Soil Characterization '.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.... 129
7.4.4 Results of 1992 and 1993 EMAP Soil Characterization 132
7.4.5 CWLSA Soil Oxidation-Reduction Potential 135
7.5 EVALUATION 135
7.5.1 Seasonal Sedimentation 135
7.5.2 Long-term sedimentation 135
7.5.3 Soil Characterization: CWLSA and EMAP Studies 140
7.5.4 Soil Oxidation-Reduction 142
7.6 FUTURE RECOMMENDATIONS 143
8.0 PESTICIDES IN WETLAND SEDIMENTS AS INDICATORS OF ENVIRONMENTAL
STRESS 145
8.1 OBJECTIVES 145
8.2 METHODS 146
8.3 RESULTS 146
8.3.1 Atrazine 146
8.3.2 2,4-D 146
8.3.3 Cyanazine 147
8.4 EVALUATION AND RECOMMENDATIONS 147
9.0 DEVELOPMENT OF NEW SAMPLING METHODS AND SAMPLING TECHNIQUES 149
9.1 DEVELOPMENT AND EVALUATION OF AN INVERTEBRATE SAMPLING
DEVICE AND A WATER-LEVEL RECORDER FOR EMAP 149
9.1.1 Introduction 149
9.1.2 Objectives 150
9.1.3 Methods 150
9.1.3.1 Objective 1 150
9.1.3.2 Objective 2 154
9.1.3.3 Objective 3 154
9.1.4 Results 155
9.1.4.1 Objective 1 155
9.1.4.2 Objective 2 160
9.1.4.3 Objective 3 160
9.1.5 Evaluation 160
9.1.5.1 Objective 1 160
9.1.5.2 Objective 2 163
9.1.5.3 Objective 3 164
9.2 HORMONAL RESPONSE TO ENVIRONMENTAL STRESS:
TECHNIQUE DEVELOPMENT 165
9.2.1 Introduction 165
9.2.2 Objectives 167
9.2.3 Methods 167
9.2.4 Results 168
9.2.5 Evaluation and Recommendations 170
10.0 RECOMMENDATIONS FOR CONTINUED USE OF INDICATORS 177
10.1 SAMPLE FRAME 177
10.2 LAND ACCESS 178
10.3 SAMPLE SIZE 179
vi
-------
10.4 SUMMARY OF INDICATOR TESTS 180
10.5 RECOMMENDED NEW APPROACH 180
11.0 PERSONNEL AND COSTS 185
11.1 EXPERTISE REQUIREMENTS 185
11.2 COSTS 186
REFERENCES 189
APPENDICES 199
VII
-------
TABLES
Table 1 -1. Physical, chemical and biological indicators of wetland condition identified for
evaluation during the 1992 and 1993 field seasons in the Prairie Pothole Region of the
United States 6
Table 2-1. Original 16 10.4-km2 plot and wetland basins (see Section 2.4) used in a pilot study
of indicators of condition of wetland basins in the Prairie Pothole Region of the
United States. All plots were used for landscape variables in 1992 and 1993. Plots
in Low-North and Low-South were dropped from the sample for all ground measurements
in 1993 13
Table 2-2, New 10.4-km2 plots and wetland basins (see Section 2.4) selected in 1993 for a pilot
study of indicators of condition of wetland basins in the Prairie Pothole Region of the
United States 15
Table 2-3. Cropland/upland ratios for 10.4-km2 plots and for drainage basins of individual sample
wetland basins used as a proxy for wetland condition 18
Table 4-1. Mission numbers and dates of photography for photographs used by the National Wetland
Inventory for mapping 4-mi2 plots used in the EMAP pilot study 26
Table 4-2. Dates on which photographs were obtained during a pilot study of indicators of wetland
condition in a pilot study of indicators of wetland condition in the Prairie Pothole
Region of the United States 32
Table 4-3. Wetland classes used for water areas that did not appear in NWI mapping 33
Table 5-1. Number of wetlands and samples per wetland needed to estimate means within 10%,
90% of the time. Note that the number of samples per wetland depends on the number
of wetlands sampled 64
Table 5-2. Mean movement (fall 1992 elevation - spring 1993 elevation) of sediment traps installed
in EMAP pilot study wetland basins during 1992. Plots 241, 246, 249, and 396 were
dropped as EMAP study sites in 1993. However, elevations were measured when we
removed equipment from basins within these plots in April, 1993 66
Table 6-1. Sampling design lay-out showing single (X or x) or multiple communities (number) sampled
within deep-marsh (DM), shallow-marsh (SM), or wet-meadow (WM) zones in basins in
good-condition and poor-condition watersheds, 1992-1993 75
Table 6-2. Response variables 34
Table 6-3. Fixed and random effects in ANOVAs 85
VIII
-------
Table 6-4. Numbers of sample wetland basins and surveyed wetland plant communities among wetland
classes in good-condition and poor-condition watersheds and mean numbers of communities
per basin, EMAP study, Prairie Pothole Region, 1992-1993 87
Table 6-5. Number and least squares means (±SE) of community areas among Stewart and Kantrud
(1971) wetland zones in good-condition and poor-condition watersheds, EMAP study, Prairie
Pothole Region, 1992-1993 88
Table 6-6. Number and total area of plant communities sampled among Stewart and Kantrud (1971)
phases in good-condition and poor-condition watersheds, EMAP study, 1992-1993 89
Table 6-7. Means (±SE) of visual estimates of proportional areas of phases of wet-meadow,
shallow-marsh, and deep-marsh zones in sample wetland basins in good-condition and
poor-condition watersheds, 1992-1993a 90
Table 6-8. Means (±SE) of proportion of zones of basins in good-condition and poor-condition
watersheds subjected to various current and recent past land uses, EMAP study, Prairie
Pothole Region, 1992-19933 92
Table 6-9. Means (±SE) proportion of major watershed cover types in good-condition and
poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993a 93
Table 6-10. Means (±SE) proportion of watersheds in current and recent past land use practices for
sample wetland basins in good-condition and poor-condition watersheds EMAP study,
Prairie Pothole Region, 1992-19933 94
Table 6-11. Means (±SE) proportion of currently raised crops on annually tilled land in watersheds in
good-condition and poor-condition watersheds, EMAP study, Prairie Pothole Region,
1992-19933 95
Table 6-12. Least squares means (±SE) water depth (cm) in plant communities3 in wet-meadow,
shallow-marsh, and deep-marsh zones of sample wetland basins in good-condition and
poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993 96
Table 6-13. Least squares means (±SE) of percent standing dead vegetation in plant communities3
in wet-meadow, shallow-marsh, and deep-marsh zones of sample wetland basins in
good-condition and poor-condition watersheds, EMAP study, Prairie Pothole
Region, 1992-1993 96
Table 6-14. Least squares means (±SE) of litter depth (cm) in plant communities3 in wet-meadow,
shallow-marsh, and deep-marsh zones in sample wetland basins in good-condition and
poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993 97
Table 6-15. Least squares means (±SE) of percent unvegetated bottom in plant communities3
in wet-meadow, shallow-marsh, and deep-marsh zones of sample wetland basins in
good-condition and poor-condition watersheds, EMAP study, Prairie Pothole
Region, 1992-1993 97
Table 6-16. Least squares means (±SE) of percent open water in plant communities3 in wet-meadow,
shallow-marsh, and deep-marsh zones of sample wetland basins in good-condition and
poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993 98
ix
-------
Table 6-17. Mean (n=5 quadrats/community) physical and vegetational features of emergent plant
communities3 in phases of zones of sample wetland basins in wetlands in good-condition
and poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993 101
Table 6-18. Total plant taxa and least squares means (±SE) taxa richness for communities in wet-meadow,
shallow-marsh, and deep-marsh zones in sample wetland basins in good-condition and poor-
condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993 102
Table 6-19. Number of plant taxa recorded in communities among phases in sample wetland basins in
good-condition and poor-condition watersheds, EMAP study, 1992-1993a 103
Table 6-20. Total numbers of perennial, annual (includes biennial), native, and introduced plant
taxa in communities in wet-meadow, shallow-marsh, and deep-marsh zones of sample
wetland basins in good-condition and poor-condition watersheds, EMAP study,
Prairie Pothole Region, 1992-1993 104
Table 6-21. Mean areal cover values and life history status in the prairie pothole region for the 10
most abundant species in communities in phases of wet-meadow zones of sample basin
wetlands in good-condition and poor-condition watersheds, EMAP study, 1992-1993.
Seeded crop plants are marked with an asterisk(*) 105
Table 6-22. Hypothetical environmental condition scores for upland 112
Table 6-23. Hypothetical environmental condition scores for low-prairie and wet-meadow zones of
prairie wetlands 113
Table 6-24. Example of an expanded ranking system for environmental condition of
wet-meadow zones 114
Table 6-25. Land use of wet meadow zones of prairie wetlands as related to environmental condition
and water levels as indicated by critical structures, functions, and stressors 115
Table 7-1. Analysis of selected chemical constituents of trapped sediments, 34 sample wetlands,
North and South Dakota, collected in 1993. Abrev.: OM = organic matter, CCE = calcium
carbonate equivalent, SD = standard deviation 130
Table 7-2. Estimation of soil loss in 4 CWLSA wetlands plus a non-eroded control site using Cs-137
analysis (Soileau et al. 1990). Bulk density values are mean values from three field
samples, Cs-137 activities are mean values of two laboratory runs. Each Cs-137 sample
was a composite of three basin subsamples. Gamma ray count time was 57600 seconds . 132
Table 7-3. CWLSA Soil nutrient analysis of 0-15 cm samples from wetlands P1, T1, P7, and C7,
Cottonwood Lake Study Area, 1992. Abrev.: WM = wet meadow, SM = shallow marsh,
NO3' = nitrate. Units: OM = % mass, NO3 and P g/m3 133
Table 7-4. MRPP statistical analysis of soil nutrient data, CWLSA, 0-15 cm soil depth, 1992 134
Table 7-5. Least squares means for nitrate and phosphorus, EMAP sample wetlands, 1992-93 136
Table 7-6. Least squares means for log transformed percent organic matter (OM) and electrical
conductivity (EC), EMAP sample wetlands, 1992-93. pH data was not log transformed ... 137
-------
Table 7-7. Least squares means for log-transferred percent sand, silt, and clay; EMAP sample
wetlands, 1993 138
Table 7-8. Soil oxidation-reduction potential measurements from 3 CWLSA wetlands.
September, 1992-June, 1993 138
Table 8-1. Atrazine concentration in wetland sediments determined by ELISA. The lower detection
limit was 15 ppb 147
Table 9-1. Legal descriptions of tracts of land containing wetlands used to evaluate quantitative
devices that sample recalcitrant remains of selected aquatic macroinvertebrates and
sediment deposits. All wetlands are located within Stutsman County, North Dakota 151
Table 9-2. Correlations with probability values (P-value) in parentheses between abundance of
invertebrate remains captured in 3 types of sediment traps (bottle-top, funnel-top,
and straight-tube) and invertebrate abundance of wetlands determined from monthly
sweep-net samples with most influential observations removed, 1993 159
Table 9-3. Correlations with probability values (P-value) in parentheses between biomass of
invertebrate remains captured in 3 types of sediment traps (bottle-top, funnel-top, and
straight-tube) and invertebrate biomass of wetlands determined from monthly sweep-net
samples with most influential observations removed, 1993 159
Table 9-4. Results of linear regressions to determine if macroinvertebrate abundance or biomass, as
estimated by the various sediment trap types, could be used to predict the percentage of
grassland remaining within each wetland's drainage basin 159
Table 9-5. Maximum and minimum water levels (cm) of wetlands P8 and P7 at the Cottonwood Lake
Study Area, Stutsman County, North Dakota, as recorded by prototype water-level recorders
developed for the EMAP pilot study 163
Table 9-6. Land use at sample wetlands. Site numbers correspond to locations on maps in Appendix
9.2.2. Land use categories are the same as those used in Section 6.0 171
Table 10-1. Summary of recommendations for indicator measurements tested during a pilot study of
indicators of wetland condition in the prairie pothole region 181
Table 11-1. Costs of the Prairie Pothole Pilot Project by activity. Costs are for calendar years
1992, 1993, and 1994 187
XI
-------
FIGURES
Figure 1-1. Map of the Prairie Pothole Region based on Mann (1974) showing strata
and sample wetland plots used in a pilot study of indicators of wetland condition ... 3
Figure 4-1. Process for deriving final vectors from polygon, line, and buffer files obtained from
the National Wetland Inventory 27
Figure 4-2. Process for creating digital files from video tapes 29
Figure 4-3. Example of final raster from video data for plot 374 after georeferencing and
resampling 30
Figure 4-4. Feature map raster for plot 374, showing area interpreted as water 31
Figure 4-5. Process used to derive digital data from 35-mm slides 34
Figure 4-6. Example of raster from high-altitude 35-mm photograph after georeferencing
and resampling 35
Figure 4-7. Process for adding vector data for features that do not appear in mapping by the
National Wetland Inventory 36
Figure 4-8. Process for creating SAS data sets from low-elevation photographs and field
delineation 38
Figure 4-9. Process used for creating SAS data sets representing drainage basins of each
sample wetland basin 40
Figure 4-10. Distribution off wetland basin density for plots classified as in good- and
poor-condition 41
Figure 4-11. Distribution of area of wetland/plot for good- and poor-condition plots 42
Figure 4-12. Distribution of mean distance from wetland basins to nearest basin between good-
and poor-condition plots 43
Figure 4-13. Distribution of shoreline development indices between good- and
poor-condition plots 44
Figure 4-14. Distribution of drained wetland basins on good- and
poor-condition plots 45
Figure 4-15. Distribution of lengths of drainage ditch per plot for good- and
poor-condition plots 45
Figure 4-16. Percent of wetland basins containing water and percent of wetland area covered
by water for each wetland basin class in 1992 47
xii
-------
Figure 4-17. Percent of wetland basins containing water and percent of wetland area covered by
water for each wetland basin class in 1993 48
Figure 4-18. Comparison of indices to change in pond between months by wetland class and
period. Asterisk indicates (P<0.05) 49
Figure 4-19. Comparisons of indices to change in area covered by water by wetland basin class
and period. Asterisk indicates (P<0.05) 51
Figure 4-20. Area of upland habitat classes/plot by condition class 52
Figure 4-21. Distribution of area of cropland/plot between good- and poor-condition plots 53
Figure 4-22. Distribution of exposed soil for good- and poor-condition plots 54
Figure 4-23. Numbers of five species of ducks per 40-km2 hexagon on good- and poor-condition
plots in 1992 and 1993. Asterisk indicates significance at 0.05 level 55
Figure 5-1. Placement of sediment traps in EMAP pilot study wetlands 61
Figure 5-2. Taxon richness (back transformed LSMs) of invertebrates captures in sediment traps
installed in EMAP pilot study wetlands, 1993. Black bars = 95% C.I 65
Figure 6-1. Wetland 374-225 (1993) showing sampled and unsampled hydrophyte communities,
location quadrats, and land-use of uplands 80
Figure 6-2. Collapsible quadrat frame 81
Figure 7-1. Map of the Cottonwood study area 123
Figure 9-1. Sampling devices tested for EMAP pilot study in 18 wetlands located in
Stutsman County, ND (A= straight-tube trap; B=flush trap; C=bottle=top trap;
D=funnel-top trap) 153
Figure 9-2. Configuration of sampling stations located on random transects
(A=straight-tube trap; B=flush trap; C=bottle-top trap; D=funnel-top trap; and
F=feldspar clay) 156
Figure 9-3. Prototype water level recorded designed for EMAP pilot study to measure water
depth fluctuations in wetland basins 157
Figure 9-4. Diagram of prototype water level recorder showing how changes in water
levels move the float and thus the indicators providing a measurement of
water level fluctuation 158
Figure 9-5. Water levels recorded with Telog water level monitor of wetland P8 at the
Cottonwood Lake Study Area, Stutsman County, ND, April to September
1992 and 1993 161
Figure 9-6. Water levels recorded with Telog water level monitor of wetland P7 at the
Cottonwood Lake Study Area, Stutsman County, ND, April to September
1992 and 1993 162
xiii
-------
Figure 9-7. Plasma corticosterone levels after low (< 6 min), medium (6-30 min), and high
(>30 min) acute stress 169
Figure 9-8. Relation between the amount of cropland surrounding wetlands on the drift plain
and the acute stress response 172
Figure 9-9. Size off tiger salamander larvae captured at different sampling periods from
wetlands occurring on the drift plain and on the coteau 173
Figure 9-10. Relation between larval size and acute stress response 174
Figure 9-11. Relation between larval size and baseline plasma corticosterone levels
(i.e., levels in larvae that were not acutely stressed) 175
Figure 9-12. Relation between acute stress response and baseline corticosterone levels 176
XIV
-------
SUMMARY
This report describes the objectives of the Environmental Monitoring and Assessment Program
(EMAP)-Wetlands. Additionally, it describes a pilot project conducted by the Biological Resources
Division-U.S. Geological Survey in the Prairie Pothole Region (PPR) of the United States to evaluate
the ability of wetland indicators to distinguish between good- and poor-condition areas. Good-condition
areas were assumed to be those least impacted by cropping practices. Thus, the good- and poor-
condition areas were based on the ratio of cropland to total area of upland, such that the smallest ratios
represented the most grassland while the largest ratios represented the most cropland. Good- and
poor-condition paired study plots were then selected from each of the three major ecoregions (Mann
Wetland Density) of the PPR (16 original plots).
The purpose of the pilot study was to select and evaluate indicators that would be robust
enough to eventually describe wetland conditions for all of the PPR or for a State via probability survey
sampling. Indicator selection involved three steps: (1) consultation with PPR wetland experts to develop
a preliminary list of indicators, (2) refinement of the list to bring it in line with budgetary and logistic
constraints, and (3) field studies to determine whether the indicators could differentiate between
landscapes in good (mostly grassland landscapes) and poor (mostly cropland landscapes) condition.
A variety of physical, chemical and biological indicators were tested during the summers of
1992 and 1993 on 12 of the original 16 plots selected. Among the physical indicators, those most
capable of differentiating good- and-poor condition wetland landscapes were: (1) frequency of drained
wetland basins, (2) total length of drainage ditch per plot, (3) amount of exposed soil subject to erosion,
and (4) indices of change in area of wetland covered by water. Among the chemical indicators tested,
only soil and sediment phosphorus conclusively differentiated between good- and poor-condition
wetland landscapes. Biological indicators included (1) invertebrates, (2) waterfowl, and (3) plant
community components. Among the various measures made for invertebrates, only taxon richness
showed promise in distinguishing good and poor conditions. Breeding pair duck counts also were
capable of distinguishing good and poor conditions. Among the plant community indicators, species
richness in the wet meadow zone was the one indicator that distinguished itself in differentiating
between good- and poor-condition wetland landscapes.
xv
-------
A major complication of conducting indicator evaluation research in this area is access to
private lands. Nearly 0.3 person-years of effort were necessary to obtain access to the 16 plots. Access
was authorized on 68% of the targeted sites. Access authorization was later rescinded on five study
sites in poor condition.
Based on the pilot study results, several indicators were recommended for further evaluation in
a follow-on study that will use probability based sample site selection.
XVI
-------
Section 1.0
INTRODUCTION
Lewis M. Cowardin
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
Spencer A. Peterson
U.S. Environmental Protection Agency
EMAP-Wetlands
Corvallis, Oregon
1.1 OBJECTIVES OF EMAP PROGRAM
The Environmental Protection Agency (EPA) initiated the Environmental Monitoring and
Assessment Program (EMAP) in 1989 to address four objectives (EPA 1993).
1. To estimate the current status, trends, and changes in selected indicators of condition
of the Nation's ecological resources on a regional basis with known statistical
confidence.
2. To estimate the geographic coverage and extent of the Nation's ecological resources
with known confidence.
3. To seek associations between selected indicators of natural and anthropogenic stresses
and indicators of ecological resources.
4. To provide annual statistical summaries and periodic assessments of the Nation's
ecological resources.
EMAP was partitioned into seven ecological resource classes:
1. agricultural 5. Great Lakes
2. rangelands 6. landscape ecology
3. estuaries 7. surface waters-lakes, streams and wetlands.
4. forests
1
-------
This report addresses wetlands. The objectives of the EMAP-Wetlands group generally parallel those of
the EMAP program except that the wetlands group will not estimate the extent of the wetland resource.
Instead, we have adopted the estimates of wetland extent and distribution reported in the U.S. Fish and
Wildlife Service's (USFWS) Congressionally mandated National Wetlands Inventory (NWI). This allows
us to focus on indicators of condition (Peterson 1994). Therefore, this report addresses a pilot project to
evaluate the performance of wetland condition indicators and their ability to discriminate between good-
and poor-condition landscapes in the Prairie Pothole Region (PPR) of the United States (See Section
2.3 for definition and selection process for good and poor landscapes).
1.2 PRAIRIE POTHOLE REGION
The PPR is situated in the northern plains of the United States and Canada. Although the
characteristics of the region have been described in detail (Kantrud et al. 1989, van der Valk 1989)
there is variation in the bounding of the region. We used a map prepared by Mann (1974) to define the
bounds because, unlike some other published maps, Mann's map covers the entire region including
Canada and delineates wetland-basin density, thus furnishing a basis for ecological regionalization. The
map has also been used as the basis of other maps and analyses that have recently been published
(e.g., Sargeant et al. 1993, Sargeant and Raveling 1992). The PPR with regions based on wetland
density is shown in Figure 1-1.
Prairie potholes are glacial in origin. They tend to be extremely variable in size, hydrology, flora,
and fauna. They are also numerous and small, which causes sampling problems unique to the region
(Cowardin et al. 1995). Climate in the PPR is unstable. The region cycles between wet and dry periods.
Spatial complexity and climatic variability in the region further confound attempts to monitor and
evaluate wetland conditions. Wetlands in this region have long been recognized as critically important
for breeding waterfowl (Smith and Stoudt 1964). More recently, society has recognized numerous other
wetland values such as water quality improvement, flood attenuation, and biological integrity (Kantrud et
al. 1989, van der Valk 1989, Peterson 1994). The region is one of the most intensively managed
agricultural areas in the United States. Agriculturally related disturbance has resulted in numerous
controversies between those interested primarily in agriculture and those interested primarily in
waterfowl production. These controversies, in turn, have resulted in numerous wetland protection laws
(Sidle 1983).
-------
USTRATUM -HIGH
Ff?a STRATUM - MEDIUM
VTA STRATUM - LOWNORTH
ESS STRATUM - LOWSOUTH
133 HAUJC POOR CONDITION
156 NORMAL GOOD CONDITION
SAMPLED 1992 -1903 GROUND AND AIR
SAMPLED 1993 GROUND ONLY
SAMPLED 1993 AIR ONLY
Figure 1-1. Map of the Prairie Pothole Region based on Mann (1974) showing strata and sample wetland
plots used in a pilot study of indicators of wetland condition.
The ecological importance of the area and the stress on the system resulting from agriculture
caused EPA to select it as one of the first areas for developing and evaluating ecological indicators of
wetland condition. Condition is defined here as the wetland state relative to the set of wetland values
defined in Section 1.3, below.
1.3 INDICATORS OF CONDITION
The condition of an ecosystem must be monitored relative to some reference. The process is
similar to that of monitoring certain parameters of body function on an individual and comparing these
measures to established norms for a healthy individual (Schaeffer et al. 1988). A variety of values is
associated with PPR wetlands. However, some of these values stand out as more significant than
others, based on the expert opinion of regional wetlands experts. Peterson (1994) and Rosen et al.
(1995) reported that the most significant values, among many, for the PPR were as follows:
-------
Biological Integrity. The sustainability of a balanced, integrative, adaptive community of
organisms having a species composition, diversity, habitat and functional organization comparable to
that of natural wetlands in the region (adapted from Karr and Dudley 1981).
Harvestable productivity. The quantity and/or quality of any service or product that wetlands
provide society (e.g., wildlife, recreation, and food production).
Water Quality Improvement. The ability of wetlands to assimilate nutrients, trap sediments, or
otherwise reduce downstream pollutant loads.
Flood Attenuation. The ability of wetlands to temporarily intercept and store surface water
runoff, thus changing sharp runoff peaks to slower discharge over longer periods of time (Mitch and
Gosselink 1986).
Peterson (1994) argued that the biological integrity value, more than any other, best defines the
reality of a reference condition, since it requires that sample site conditions be compared with those of
least impacted wetlands in the region. Also, because of this requirement, biological integrity represents
a set of conditions more basic and less disturbed by human activity, compared to the other three
values. Indeed, the other three values are nearly always managed for improvement. We recognize that
few if any undisturbed wetlands, and thus true reference conditions, exist in the PPR. Thus
determination of reference conditions is dependent on our ability to select meaningful ecological
indicators and measure them over the range of their existing conditions, assuming that those of highest
quality represent a reasonable reference condition. Another approach is to select a biased sample of
"good-" and "poor-" condition sites based on our preconceived notion that certain readily identified
factors (agricultural practices) contribute to the degradation or enhancement of wetland conditions.
Ecological indicators capable of discriminating between the good- and poor-conditions should be useful
not only in describing reference conditions (good sites), but also the range of conditions that might be
encountered if probability sampling of the entire region were conducted. With the second approach,
condition could be determined as a cumulative distribution function for an indicator or group of
indicators as described by Overton et al. (1990).
For this pilot study, we chose to designate good- and poor-condition sites (defined in terms of
cropland as explained in Sections 2.2 and 2.3). Ecological condition indicators were selected relative to
their ability to address the values above (biological integrity, harvestable productivity, water quality
-------
.improvement, and flood attenuation). Selected indicators are shown in Table 1-1. Indicators were then
tested for variance and for their ability to discriminate between our predefined extremes in condition
(good and poor). The idea behind this approach, besides being dictated by budget level, was that
indicators incapable of distinguishing between good and poor extreme conditions would be of little use
in distinguishing among several intermediate conditions likely to be encountered when probability
sampling accessed the full range of conditions. Thus, while the good/poor landscape condition
evaluation approach for indicators was taken in the pilot project, it was with the idea of refining the list
of indicators for use over the entire range of wetland conditions that might be encountered in the PPR.
The pilot study was designed both to begin setting the frame of reference for ecosystem condition and
to select a meaningful and practical set of indicators to be used during probability sampling. The
rationale and approach to the pilot is more completely described in Section 1.4, below.
1.4 RATIONALE FOR THE PILOT STUDY
The purpose of the pilot study was to select and evaluate indicators that would be robust
enough to eventually describe wetland conditions for a large region (all of the PPR or a State) using a
probability sampling design. The process involved three steps: (1) consultation with experts on the PPR
to decide on a preliminary list of potential indicators, (2) refinement of that list to bring it in line with
budgetary and logistic constraints, and (3) field studies designed to determine whether the selected
indicators do differ for sites that are highly disturbed (those with a high ratio of cropland to upland) and
sites that are not (having a low ratio of cropland to upland). An indicator incapable of distinguishing
extremes of condition would be of little use in attempting to distinguish less extreme conditions likely to
be encountered during probability sampling. The site selection process for this pilot study is different
from the methodology that eventually will be used to characterize condition of the entire region for
probability sampling because the test sites in this study were hand-picked based on the ratio of
cropland to upland definition above. Thus, no estimates for the population of all wetlands in the region
can be made from the data contained in this report.
Probability sampling, on the other hand, draws samples randomly from the universe of wetland
basins in the region and can be used to extrapolate wetland condition estimates for the entire region.
Another advantage of probability sampling is that it can also can provide confidence limits for the
condition estimates. Because probability sampling is expected to encounter every condition imaginable,
however, it is critical that we understand how any particular condition indicator performs under the best
and worse of conditions.
-------
Table 1-1. Physical, chemical and biological indicators of wetland condition identified for evaluation
during the 1992 and 1993 field seasons in the Prairie Pothole Region of the United States.
Category
Indicator
Physical Landscape
Soil/Sediment
Water Chemistry3
Biological Indicators
Invertebrates
Amphibians
Waterfowl
Plant Community Indicators
Wetland basin density
Area of wetland
Shoreline development indices
Drained wetland basins
Lengths of drainage ditches
Percent of wetland basins containing water
Percent of wetland area covered by water
Seasonal change index by wetland class
Change in area index by wetland class
Area of upland habitat class
Area of cropland by landscape class
Exposed soil subject to erosion
Soil class
Nitrate-nitrogen content
Phosphorus content
Organic matter
Conductivity
Salinity
PH
Particle size
Taxon richness
Biomass
Abundance
Sedimentation
Salamander acute stress
Duck counts
Number of community areas by wetland zone
Wetland zone types (%)
Land use types abutting wetland (%)
Watershed cover in annuals and perennials (%)
Standing dead vegetation by zone (%)
Litter depth by zone
Unvegetated bottom in plant community (%)
Total plant taxa by richness and zone
1 Dropped midway through 1992 due to severe drought, i.e., little if any water to sample in many wetlands.
6
-------
EPA convened a meeting of scientists and managers experienced in prairie wetland
ecosystems at the United States Geological Survey's Northern Prairie Science Center in July 1991.
Individuals attending the meeting are listed in Appendix 1-1. The meeting had four objectives
1. to review what is known about the ecology of the prairie wetlands and the research and
monitoring efforts currently being conducted
2. to discuss Federal research programs designed for the prairie potholes
3. to identify areas of cooperation
4. to initiate a process to refine an approach for monitoring the condition of the northern
prairie wetlands.
The meeting resulted in agreement on two underlying characteristics of prairie potholes, critical to
planning an effective monitoring program. First, climate of the region cycles between extremes from wet
to dry and is the primary factor determining the characteristics of prairie potholes. Therefore, a program
for research and monitoring must be long-term to address this type of temporal variation. Second,
wetlands within the prairie pothole region show great spatial variation in climate, geology, hydrology,
fauna, and land use.
This study evaluates the performance of selected wetland condition indicators by measuring
their responses at good- and poor-condition sites as described above (ratio of cropland to upland). Our
indicators of condition are strongly correlated with landform. We use ecoregions based on landform to
help overcome the confounding effect of landform. Previous studies have used ecoregions based on
landform to account for spatial variation (e.g., Lake Agassiz Plain, Drift Plain, Missouri Coteau; see
Stewart 1975). A primary sampling unit must be large enough to represent the wetland basin sizes and
hydrologic functions occurring in that unit. At this meeting, we listed drainage, sedimentation, altered
hydrology, proximity to cropping, effects of herbicides, pesticides, and fertilizers, burning, human
disturbance, and livestock as the principle stressors of prairie wetlands. Therefore, we assumed that
intensive agriculture degrades wetland condition, and wetlands in relatively undisturbed areas are
functionally in better condition than those in areas of intensive agriculture. Because of the dynamic and
cyclic nature of the prairie pothole system, definition of condition and measurement of changes or
trends in condition is more difficult than in other, more predictable, wetland systems and will require
sampling over a long period of time.
-------
A preliminary list of potential indicators of condition included land use, percent of wetland
basins containing water each year, water chemistry, community composition and abundance of
vegetation, sedimentation rate, and macroinvertebrates. The meeting attendees agreed that final
selection of a list of potential indicators would require further evaluation, because no robust data sets
representing statistically defined variability of indicator response over a wide range of conditions existed
for the region.
A second meeting was held at NPSC in September 1991 between EMAP and NPWRC
administrators and potential principal investigators (Appendix 1-2). It resulted in a refined list of
indicators to be tested in the pilot study. The list was constrained both by practicality of making certain
measurements during the narrow time frame of a 2-year pilot study and by availability of funds and
personnel. The list included landscape indicators, hydrology, sediment characteristics, vegetation
composition and abundance, faunal composition and abundance, water and soil chemistry, chemical
contaminants, and stressor information. Further refinement of the list of indicators occurred during
development of a plan of work. For example, chemical contaminants were dropped because of the cost
of analysis even though their importance as an indicator of condition was stressed at both planning
meetings. Other indicators such as shorebird populations proved impractical because of time, personnel
and funding constraints. After the first year of the study it became obvious that any indicator that
required water quality measurements such as electrical conductance, pH, and chlorophyll as well as
faunal measurements such as number of amphibians were impractical because the vast majority of the
wetland basins were dry. These indicators were dropped as detailed in the individual studies that follow.
A final list of indicators for the pilot study was listed by Dwire (1994).
1.5 OBJECTIVES OF PILOT STUDY
The functions of individual prairie potholes are intimately related to other potholes and to the
upland matrix that contains those potholes. The pilot study, reported here, had three objectives
1. to test selected landscape and field indicators of condition by discriminating between
wetlands in highly disturbed (agricultural) landscapes and those in least disturbed
(grassland) landscapes across the U.S. portion of the PPR
2. to develop new indicators and refine sampling techniques for prairie potholes at a
subset of the sites and reference areas
-------
3. to identify and explore resolution of issues related to access to private land and
logistics.
1.6 ORGANIZATION OF THE REPORT
This report is organized into a number of sections that represent work by individual researchers
whose names appear with the section for purposes of citation. Section 2 describes the overall design
for studies intended to meet Objective 1 where various studies were conducted on the same wetland
basins within the same plots. Details of sampling within the basins as well as descriptions of study
areas selected for testing of methodology are incorporated as appropriate in Sections 4 through 8.
Section 3 deals with logistics and landowner access problems (Objective 3) encountered during sample
selection and conduct of studies for plots described in Section 2. Section 4 reports on landscape level
indicators where the plot rather than the wetland basin was the sampling unit. Section 5 reports on
invertebrates as an indicator of wetland condition. Data were derived from newly-designed sampling
devices installed in the sample wetland basins. Sections 6 and 7 report on vegetation and soils
indicators that were measured on the same wetland basins and at the same time. Section 8 describes
pesticide residues found in soil samples gathered during field work described in Sections 6 and 7.
Section 9 reports results for tests of techniques (Objective 2) developed for use in measuring indicators
of condition. For logistical and design reasons, this work was conducted at 18 sites in Stutsman
County, ND, most of which were in Waterfowl Production Areas so that access would not be a problem,
rather than at the pilot study sites described in Section 2. Section 10 makes general recommendations
to EPA and summarizes results for the various indicators tested during the pilot. Section 11
summarizes costs of the pilot study for possible use in planning future work.
-------
Section 2.0
DESIGN METHODS
Lewis M. Coward! n
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
The studies designed to accomplish Objective 1 (see Section 1.1) for the test of indicators of
wetland condition for the Prairie Pothole Wetlands described in Sections 4 through 8 of this report
shared a common sampling plan, which is described below.
2.1 REGIONALIZATION
Prairie wetlands vary by geographic region. To minimize the confounding effect of regional
differences on our comparison of landscapes that are highly disturbed from those that are not, we
stratified the Prairie Pothole Region (PPR) into four ecoregions of high, medium, and low wetland basin
density based on Mann's (1974) map (Fig. 1-1). The high-density ecoregion is approximately equivalent
to the Missouri Coteau, a large morainal belt trending northwest to southeast and characterized by
collapsed hummocky topography "dead-ice moraine" (Bluemle 1991). The medium-density ecoregion is
approximately equivalent to the drift plain, an area of glacial drift with less relief lying east of the
Missouri Coteau. We divided the low-density region into north and south regions. The Low-North
ecoregion represented the Red River Valley, which is composed of the bed of glacial Lake Agassiz.
The Low-South ecoregion is similar to the medium region except that most of the wetlands have
already been drained.
2.2 STUDY PLOTS
There was strong consensus at the planning meetings that assessing the condition of prairie
wetlands would require an assessment of both wetland complexes and the uplands that surrounded
them. Unfortunately, there was no usable definition of what constituted a complex. During the second
planning meeting the attendees decided that all wetlands within a 40-km2 hexagon sampling unit (or
"hexagons") as defined by EPA (Overton et al. 1990) could serve as a wetland complex. Experts
attending the meeting on the PPR were also unable to define wetland condition, but they did agree that
11
-------
wetlands in a complex containing predominantly cropland would probably be in a more degraded state
than those containing predominantly grassland. Therefore, we proposed the ratio of cropland area to
total upland area in each hexagon as a proxy for wetland condition and decided to evaluate indicators
by their ability to distinguish between wetlands at the extremes of that proxy.
To develop the ratio, we needed a list of wetland basins (e.g., Stewart and Kantrud 1971) in
each basin class from which to draw sample sites as well as estimates of the area of upland land
covers for determining the cropland/upland ratio. No such data existed for EMAP 40-km2 hexagons,
although they were needed immediately to prepare study plans and begin field work in the spring of
1992. Therefore, we decided to use an existing sample of 10.4-km2 (4-mi2) plots rather than the 40-km2
hexagons. A large sample of these plots (422) had previously been selected to furnish data for a
mallard simulation model (Cowardin et al. 1988). This was acceptable because our primary purpose
was to evaluate indicators of condition, not to make statements about the population of wetlands in the
region. Wetlands on each 10.4-km2 plot were mapped according to the classification of Cowardin et al.
(1979), and uplands were mapped according to a simplified classification that included grassland and
cropland (Cowardin et al. 1988). All map data had been digitized into the Map Overlay and Statistical
System (MOSS) and Statistical Analysis System (SAS) files that described all polygons on the maps
that had been prepared (Cowardin et al. 1988). Detailed procedures used in processing of National
Wetlands Inventory (NWI) data are described in section 3.2.1.
2.3 METHOD OF SELECTING PLOTS TO REPRESENT EXTREMES IN
CONDITION (CROPLAND/UPLAND RATIO)
Our intent was to select plots with a maximum spread between those that contained mostly
cropland (poor-condition) and those with mostly non-cropland (good-condition). For each available plot,
we calculated the ratio of area of cropland to the total area of upland. The smallest ratio represented
the most grassland and the largest ratio the most cropland. Next the plots were sorted by the ratio
within ecoregion. We then selected four plots in each ecoregion, the two with the highest ratio of
cropland (poor-condition) and the two with the lowest ratio of cropland (good-condition). Our sample
size was constrained by funding level rather than statistical considerations. We also constrained plot
selection by the following rule: Each pair of plots (the one with the lowest cropland ratio and the one
with the highest cropland ratio) must contain at least two temporary, two seasonal, and two
semipermanent wetland basins. This selection process maintained separation between high- and low-
cropland ratio landscapes in all ecoregions except the Low-North where there were few available plots
12
-------
and few semipermanent wetland basins. In that region it was not possible to maintain a meaningful
separation, and the requirement for having semipermanent wetland basins was dropped. The final
selection process resulted in the original 16 plots (Table 2-1).
Table 2-1. Original 16 10.4-km2 plot and wetland basins (see Section 2.4) used in a pilot study of
indicators of condition of wetland basins in the Prairie Pothole Region of the United
States. All plots were used for landscape variables in 1992 and 1993. Plots in Low-North
and Low-South were dropped from the sample for all ground measurements in 1993.
Wetland Density
Region Condition
Low North Poor
Poor
Poor
Poor
Good
Good
Good
Good
Low South Poor
Poor
Poor
Poor
Poor
Poor
Good
Good
Good
Good
Good
Good
Medium Poor
Poor
Poor
Poor
Poor
Poor
Poor
Plot
38
38
54
54
59
59
60
60
246
246
246
246
241
241
249
249
249
396
396
396
134
134
134
134
134
134
145
Basin
Class
Temporary
Seasonal
Temporary
Seasonal
Temporary
Seasonal
Temporary
Seasonal
Temporary
Temporary
Seasonal
Semipermanent
Seasonal
Semipermanent
Temporary
Seasonal
Seasonal
Temporary
Semipermanent
Semipermanent
Temporary
Temporary
Seasonal
Seasonal
Semipermanent
Semipermanent
--
Wetland
Basin
44
62*
39
24*
42
111
128
58
34
37
52*
53
48
3
50
72
86
107
106
130
270
432
158"
406
140
165"
—
13
-------
Table 2-1. (continued)
Wetland Density
Region Condition
Good
Good
Good
Good
Good
Good
High Poor
Poor
Poor
Poor
Poor
Poor
Poor
Good
Good
Good
Good
Good
Good
Plot
363
363
374
374
374
374
441
442
442
442
442
442
442
73
73
156
156
156
156
Basin
Class
Temporary
Semipermanent
Temporary
Seasonal
Seasonal
Semipermanent
--
Temporary
Temporary
Seasonal
Seasonal
Semipermanent
Semipermanent
Temporary
Semipermanent
Temporary
Seasonal
Seasonal
Semipermanent
Wetland
Basin
58
22
65
225
272
100
--
260
261
93
281
295
301
86
29
26
24
42
22
'Permission (or access to these plots was rescinded.
bReplaced wetland basin 272, which was drained during the study.
"Replaced wetland basin 193 for which permission was denied prior to field work.
Changes to this sample design became necessary when we were refused access to certain
wetland basins by some landowners. When refusal resulted in inability to obtain samples from the two
wetland basins in each class of wetland basins in each extreme pair of plots, we selected the next most
14
-------
.extreme plot, resampled wetland basins and again contacted landowners. The process was repeated
until the sample was complete and our original constraint was met. We refer to a pair of good- or poor-
condition plots within an ecoregion (e.g. Low-North) as a stratum. The sample was further altered at
the end of the 1992 field season because drought and access denials resulted in almost no data from
the low wetland-density strata. Those strata were dropped from the sample and we selected two new
samples from the high and the medium wetland-density strata (Table 2-2).
Table 2-2. New 10.4-km2 plots and wetland basins (see Section 2.4) selected in 1993 for a pilot study
of indicators of condition of wetland basins in the Prairie Pothole Region of the Unites
States.
Wetland Density
Region
Medium
High
Condition
Poor
Poor
Poor
Good
Good
Good
Poor
Poor
Poor
Good
Good
Good
Plot
133
133
133
498
498
498
327
327
327
407
407
407
Basin
Class
Temporary
Seasonal
Semipermanent
Temporary
Seasonal
Semipermanent
Temporary
Seasonal
Semipermanent
Temporary
Seasonal
Semipermanent
Wetland
Basin
370a
366
3BO
227b
277
146
72
147
117
109
67
168
"Replaced wetland basin 27 because permission was rescinded.
"Replaced wetland 227 which was an error in NWI data.
2.4 SELECTION OF SAMPLE WETLAND BASINS WITHIN PLOTS
We classified the wetland basins (Cowardin 1982) using a modification of Stewart and
Kantrud's (1971) classification.
15
-------
We first grouped all wetland polygons (mapping units, Cowardin et al. 1988) classified in the
data set according to Cowardin et al. (1979) into basins. The polygon with the most permanent water
regime was then used to determine the Stewart and Kantrud (1971) pond class (equivalent to wetland
basin) for the group of polygons included in the basin.
Each wetland basin, in each wetland basin class (i.e., temporary, seasonal, semipermanent), in
each ecoregion was assigned a random number. A sorted list of these random numbers was prepared
for the pairs of plots representing good- and poor-condition. The top two wetlands in each wetland
basin class were selected with the following constraints:
1. Reject all wetlands mapped as linear or point features. When the maps were
constructed, those wetlands that were too narrow or too small to enclose with a polygon
were mapped as linear or point features. The linear wetlands were almost all ditches
used to drain wetland basins in agricultural fields. Point wetlands were almost all
dugouts constructed for watering stock. EPA did not wish to include these highly
artificial entities in an evaluation of indicators of condition.
2. Reject all wetland basins containing a dugout. These wetlands, though mapped as
polygons, were also highly abnormal.
3. Reject all basins containing lacustrine wetland. These lakes are highly variable and
would require increases in sample size to avoid confounding a comparison of wetland
condition. Furthermore, nearly all of them appear on 1:100,000 USGS quadrangle maps
and will be included in EMAP's lake surveys.
4. Reject all temporary and seasonal wetlands that are not completely within the
plot boundary. Although, this procedure biases the selection procedure, it was
unavoidable because we had no data outside the plot boundary. For multi-polygon
basins it was not possible to classify the basin. We included portions of semipermanent
basins that were partly within the plot because the basin class was known by default
and exclusion would bias the sample against large basins. For the purpose of the pilot,
bias in selection of the study basins is not as important as assuring that they will satisfy
the objectives of the pilot and EMAP.
16
-------
5. Reject wetland basins in good-condition plots if the basin is surrounded by
cropland. Cropland does occur on plots that have the best condition available. At a
landscape scale these plots may be adequate, but a site-specific wetland basin
surrounded by cropland does not furnish a good comparison to a similar wetland basin
in a poor-condition plot.
6. Reject wetland basins in poor-condition plots if the basin is surrounded by
upland other than cropland. This is the corollary of criterion 5.
2.5 SELECTION OF REPLACEMENT WETLAND BASINS AND PLOTS
After selection of plots and wetland basins, each landowner was contacted to obtain permission
to enter land for sampling (see Fellows and Buhl 1995 for details). If we were refused permission, that
wetland basin was dropped from the sample and the next wetland basin of the same class was
obtained from the list of basins that had been sorted by random number. We repeated the process until
we were granted access to the land. When rejection of access resulted in no available wetland basins
of the proper class required to meet our plot selection criteria, we were forced to reject the entire plot
and draw the next plot from the list of plots that had been ranked by cropland/upland ratio. The problem
of rejection continued into the field season, when some landowners rescinded permission. In these
cases we sometimes had already gathered part of the data for the plot. When this happened, the
remainder of the data were gathered from the nearest available wetland basin of the same class. The
problem caused data for different indicators to be collected at different sites in a few cases.
2.6 REASSIGNMENT OF WETLAND BASIN CONDITION
When field work was started, the only data available for assignment of condition to individual
wetland basins were from old aerial photographs for the plots. The condition definition for the individual
basins used in the vegetation and soil studies was taken directly from the definition for the plot that
contained them. In some cases this definition proved misleading as evidenced by data gathered during
the studies. There were three reasons for this
1. classification errors in the data from the original 422 plots used in the selection process
17
-------
2. major landscape changes that occurred since the original photographs were taken, the
most important being addition of Conservation Reserve Program (CRP) cover
3. individual wetland basin condition that was more closely related to the uplands in the
surrounding drainage basin than the condition of uplands on the entire plot.
For purposes of analysis, we decided to reassign wetland basin condition, because we now
had current data on uplands and a delineation of drainage basins of the sample wetland basins (based
on aerial/video and field data, see Section 4.2.1). The new assignment was based on the
cropland/upland ratio for the individual sample wetland basins (Table 2-3). For analysis of vegetation
data, CRP cover was not treated as cropland because the new cover is more like grassland than the
tilled soil of cropland. For analysis of soils data, CRP cover was treated as cropland because the soil
parameters measured were mostly the result of runoff into the basin prior to establishment of the CRP
cover.
Table 2-3. Cropland/upland ratios for 10.4-km2 plots and for drainage basins of individual sample
wetland basins used as a proxy for wetland condition.
Plot
number
038
054
059
059
060
060
073
073
133
133
133
134
134
134
134
134
134
134
145
Wetland Basin
number
62
39
111
42
128
58
29
86
370
380
386
140
158
165
270
272
406
432
Plot
C/U ratio
0.94639
0.96792
0.54032
0.54032
0.48379
0.48379
0.00000
0.00000
0.94953
0.94953
0.94953
0.89629
0.89629
0.89629
0.89629
0.89629
0.89629
0.89629
0.96662
Basin
C/U ratio
0.78094
1.00000
0.00000
0.79960
0.00000
0.00000
0.00000
0.00000
1.00000
0.87061
0.42945
1.00000
1.00000
1.00000
1.00000
1.00000
1.00000
1.00000
0.00000
18
-------
Table 2-3. (Continued)
Plot
number
156
156
156
156
241
241
246
246
246
249
249
327
327
327
363
363
374
374
374
374
396
396
396
407
407
407
441
442
442
442
442
442
442
442
498
498
498
Wetland Basin
number
22
24
26
42
3
48
34
37
53
50
86
117
147
72
22
58
100
225
272
65
106
107
130
109
168
67
260
261
281
295
301
93
93
146
227
277
Plot
C/U ratio
0.00000
0.00000
0.00000
0.00000
0.84595
0.84595
0.92392
0.92392
0.92392
0.75344
0.75344
0.88660
0.88660
0.88660
0.04516
0.04516
0.04112
0.04112
0.04112
0.04112
0.56793
0.56793
0.56793
0.03637
0.03637
0.03637
0.96483
0.87462
0.87462
0.87462
0.87462
0.87462
0.87462
0.87462
0.10995
0.10995
0.10995
Basin
C/U ratio
0.00000
0.00000
0.00000
0.00000
0.98456
1.00000
1.00000
1.00000
0.85218
0.01821
0.35356
1.00000
1.00000
1.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.00000
0.35792
0.00000
0.04200
0.00000
0.00000
1.00000
1.00000
1.00000
0.00000
0.96514*
0.17024
0.00000
0.00000
0.00000
0.00000
"Plot contained CRP cover which was treated as Grassland for the Soils study, resulting in a ratio of 0.21879.
19
-------
Section 3.0
ACCESS TO PRIVATE LAND AND LOGISTICS
David P. Fellows and Thomas K. Buhl
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
Problems of access to private land are of critical importance in future planning for EMAP-
wetlands in the PPR. The same types of problems must be dealt with by other agencies like the
National Biological Service, which might be attempting to gather statistically valid ecological survey
data. Because of the immediacy and breath of this problem we submitted the results from this access
study to the journal Wetlands, and it has been published (Fellows and Buhl 1995). The following
summarizes the Wetlands article.
3.1 RESEARCH ACCESS TO PRIVATELY OWNED WETLAND BASINS IN THE
PRAIRIE POTHOLE REGION OF THE UNITED STATES
We attempted to obtain access for research to 81 wetland basins on 69 farms in 4 zones of the
Prairie Pothole Region of North Dakota, South Dakota, and Minnesota. We were permitted access to
54% of the farms in areas where land was intensively cropped and 87% of farms in areas of low
cropping intensity. On average, we had to contact 1.35 operators and conduct 1.70 interviews for each
successful decision.
Blanket access was not usually given--on 77% of the farms cooperators placed at least one
restriction on access. The most common restrictions were walking access only or notification before
nighttime work. No cooperators were willing to sign written access agreements.
The cost of obtaining access averaged $265/farm in wages and travel expenses.
3.2 IMPLICATIONS FOR FUTURE PROJECTS
In addition to the cost and time required to gain access, there were two other problems that
have broad implications for the type of research we intended to pursue. First, we were unable to assure
21
-------
permanent access to sites. Operators rescinded access to four farms and drained three wetland basins
during the first year; six of the seven sites lost were in the intensively cropped portion of a low-wetland-
density zone. The difficulty of obtaining and retaining research access to privately owned wetland
basins in intensively cropped areas may be related to landowner attitudes towards wetlands. We
hypothesize that farmers with crops on proposed sites are distrustful of our purposes, suspecting that
we may find conditions that would lead to further regulation. To solve this problem, researchers may
have to rely on remote sensing or consider payment for access to secure representative research sites
in such areas.
The second problem stems from changes in the law governing research projects like the Pilot
Test of Indicators of Wetland Condition for Prairie Pothole Regions. In 1993, most biological research
functions in the Department of Interior were consolidated into the newly formed National Biological
Service (NBS). Congress has now made it mandatory that NBS obtain written permission for access
from the property owner. In our pilot study, no farm operators or owners were willing to give written
permission despite our offer of a form that they could annotate to include any restrictions they desired.
The pilot test was initiated before the requirement for written access, thus it was not affected by the
operator/owners denial of written permission. However, this does appear to be a potential problem for
future surveys. Unwillingness of cooperators to sign access agreements may jeopardize research by
the newly formed NBS and other resource management agencies.
22
-------
Section 4.0
TESTS OF SELECTED LANDSCAPE INDICATORS
OF WETLAND CONDITION
Lewis M. Cowardin and H. Thomas Sklebar
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
Prairie wetlands are interrelated both in their function and in their values. In addition, their
functions and values are dependent on the surrounding uplands. An evaluation of wetland condition,
therefore, must be at the landscape as well as the individual basin level. Landscape indicators are
those that refer to an entire wetland complex and its surrounding uplands. The indicators described
here are applied at the level of primary sampling units (the 10.4-km2 plots used in the pilot as a
substitute for 40-km2 hexagons). We describe three different types of indicators: (1) landscape features
that are relatively stable, such as the number of wetland basins; (2) landscape features that vary
temporally, such as the amount of surface water in wetland basins; and (3) numbers of birds that are
not confined to a wetland basin and that use wetlands and uplands throughout the landscape.
This study had two goals: to furnish the data essential to monitoring the condition of the
integrated prairie wetland landscape and to characterize the landscape features essential to
understanding and interpreting the basin-by-basin evaluations conducted for other indicators
incorporated in the pilot. In addition, this chapter evaluates various landscape-level indicators of
condition with regard to EMAP's long-term need for indicators that can be measured at numerous sites
with minimal cost and that discriminate between wetland systems in good- and poor-condition.
Landscape level variables can be measured at a variety of scales including continental,
regional, local, and site specific. In most cases data for measuring landscape variables are most easily
derived from remote sensing. The resolution of remote sensing data can be matched to the appropriate
scale. For example at the continental scale Advanced Very High Resolution Radiometer (AVHRR) data
with a resolution of 1 km may be adequate (Loveland et al. 1991). At the regional scale Thematic
Mapper (TM) data have been effectively used for assessing wetland characteristics (Koeln et al. 1986).
This study required landscape variables at a local scale and because of the small size of prairie
potholes, at high resolution. In addition, data were required in narrow time frames which frequently can
not be obtained from satellite data. Two data sources filled these needs. Mapping techniques
23
-------
developed by the National Wetland Inventory (NWI) furnish the resolution and spatial accuracy required
for this study (Pywell and Niedzweadek 1980). Aerial video techniques (Cowardin et al. 1979, Sidle and
Ziewrtz 1990) furnish both the resolution and temporal requirements. Equipment and software for the
latter technique was in place at Northern Prairie Science Center and data sets created from previous
work were available for immediate use.
The abundance and distribution of vertebrate animal populations furnish a measure of
landscape condition for those animals that are not confined to a single wetland basin. During planning
meetings a number of species were considered as possible candidates for measurement. Counts or
indices to species abundance conducted throughout the entire area of each sample plot would be time
consuming and expensive. For purposes of the pilot, we elected to use estimates of breeding
population of five species of dabbling ducks (mallard, Anas platyrhynchos; gadwall, Anas strepera; blue-
winged teal, Anas d/scors; northern shoveler, Anas clypeata; and northern pintail, Anas act/fa).
Population estimates for these species are made each year on each 10.4-kmz plot by the USFWS and
were available for use at no cost.
Some of the measurements and preparatory work required only one visit to a site. The tasks
requiring one visit include the original mapping of the study plots and the determination of the
watersheds for sample wetlands within the plots. One-time measurements requiring an appreciable
initial effort are justified by the objectives of EMAP-Wetlands to "Quantify the regional status of
wetlands, ..." (see Leibowitz et al. 1991:2). Although there was a cost to the project for this preparatory
work, such costs would not be on an annual basis once the actual EMAP monitoring is in place.
Landscape level measurements that were unlikely to show a difference between good- and
poor-condition environments during the short span of the pilot study were excluded from the pilot study.
For example, total loss of wetlands through drainage must be measured during the long-term
monitoring of EMAP, but the drainage rate is probably so low that it is not measurable in a two-year
study. This pilot study was intended to test the technology that is practical and to demonstrate
landscape-level indicators that may separate good- and poor-condition environments in any year.
4.1 OBJECTIVES
1. Compare the spatial distribution, density, area, and characteristics of wetlands and
uplands between landscapes with high and low abundance of cropland
24
-------
2. Determine the seasonal loss of surface water in temporary, seasonal, and
semipermanent wetland basins from April through July and compare these rates
between landscapes with high and low abundance of cropland
3. Determine the extent, type, and distribution of both wetland and upland plant
communities and to relate these variables to indicators of wetland condition measured in
other sections of this report
4. Compare the size of breeding populations of 5 species of dabbling ducks using
10.4-km2 plots in high and low cropland landscapes and to relate these estimates to
landscape habitat variables including wetland distribution, upland habitat classes, and
size and juxtaposition of cover patches
5. Map the watershed of each wetland basin selected for ground study to relate
sedimentation rates to land use and management in support of soils investigations.
4.2 METHODS
4.2.1 Base Mapping of 10.4-km2 Plots.
The 10.4-km2 plot data were derived from 1:63,360 color-infrared photographs taken during the
late 1970s and early 1980s (Table 4-1) prior to completion of operational mapping by the National
Wetland Inventory (NWI). After plot selection, the data for the selected plots were updated and a
number of topological and classification errors were corrected (Fig. 4-1). One plot (134) had areas
where data were missing. Data for these areas were added by scanning current NWI maps and adding
vectors to the plot data. Wetlands were classified according to Cowardin et al. (1979) and upland
classification was after Cowardin et al. (1988). Some road linears had topological errors and were
discontinuous or had missing data. These errors were corrected prior to buffering. Three files containing
polygon buffered lines and points for wetland features and lines for non-wetland features were delivered
to NPWRC in MOSS format by NWI.
A number of processing steps were used to create a vector map file for each plot representing
current (1993) conditions. The MOSS data were imported into MIPS and *.RVF files were created for
25
-------
Table 4-1. Mission numbers and dates of photography for photographs used by the National Wetland
Inventory for mapping 4-mi2 plots used in the EMAP pilot study.
Plot Mission Date
38
54
59
60
73
133
134
145
156
241
246
249
327
363
374
396
407
441
442
498
82-062
79-068
79-068
79-068
79-057
81-046
81-046
79-068
79-057
80-046
80-046
80-049
79-056
82-061
82-061
M2959
82-062
80-046
81-046
H456
04-22-82
06-04-79
06-04-79
06-04-79
05-16-79
04-05-81
04-05-81
06-04-79
05-16-79
05-02-80
05-02-80
05-06-80
05-15-79
04-20-82
04-20-82
unknown
04-22-82
05-02-80
04-05-81
each plot. Roads and some areas classed as odd areas were mapped as linears by the NWI. For our
purposes all features in the data set had to have an area. Polygons were created by buffering points
and linear features. The NWI buffered odd areas and wetland features prior to delivering vector files to
NPWRC but did not buffer roads. We double buffered road linears to create area for the road surface
and for the right-of-way.
Photographs and video gathered during the pilot study showed wetlands that had been missed
in the original mapping as well as new wetlands that had been created since the original mapping.
These areas were digitized and added to the vector data for each plot. Although we were able to
identify new or previously missed wetlands, we were unable to classify them. They were assigned a
wetland class of UK for unknown. We also updated the upland data by adding areas that had changed
from cropland to grass-legume cover planted in response to the Conservation Reserve Program (CRP).
26
-------
FINAL VECTORS
EMAPPOLY.RVF
S»» Appendix 4-1 tordmto
Figure 4-1. Process for deriving final vectors from polygon, line, and buffer files obtained from the
National Wetland Inventory.
27
-------
Data for the location and delineation of CRP cover was obtained from county offices of the Agricultural
Stabilization and Conservation Service (ASCS).
The final result of these corrections and additions were 16 vector data sets, one for each of the
10.4-km2 plots used in the pilot study (Appendix 2-2). These data sets were current in 1993 and have
all features represented by polygons rather than lines and points.
4.2.2 Aerial Video
To evaluate the hydrologic function of wetland basins and as an aid to interpreting biological
measures of condition, we determined change in the amount of surface water in each wetland basin. To
do this we used aerial video obtained during the first week of April, May, June, and July in each year of
the study. Video (VMS format) was taken from light aircraft (Cessna 185) equipped with a belly camera
port. We used a Cohu 4810 monochrome camera equipped with a 8.8-mm charge-coupled detector, a
5.9-mm wide angle lens. The camera was equipped with a near infrared (0.81-0.89 urn) bandpass
interference filter and a Kodak Wrattan No. 0.60 neutral density filter. The signal from the camera was
recorded on a Panasonic AG-2400 portable video cassette recorder. The image was monitored in the
aircraft with a Panasonic CT-500V 14-cm color monitor. Video was obtained in two 1-mile wide swaths
at an elevation of approximately 1,829 m above ground level (AGL).
The video data were converted to a final raster file by the process illustrated in Fig. 4-2. The
process included many processing steps incorporated in MIPS software as well as hand operations.
Images representing each section (2.59 km2) of the 10.4-km2 plot were captured as separate *.RVF
files. On some plots, weather conditions forced us to fly at lower than planned altitude, so that more
than four images were required to cover an entire 10.4-km2 plot. These raw images contained
distortions caused by the attitude of the aircraft and spherical distortion of the short camera lens. We
georeferenced the images by obtaining data from the final vector data for each plot and manually
aligning the vector to the raster by means of features such as roads that appeared in both data sets.
The raster was then resampled to remove distortions. After georeferencing and resampling the
individual images for each plot, we combined them into a single raster, representing the video on each
date for each plot (Fig. 4-3).
28
-------
( VIDEOTAPE 1
Q
4
CAPTURETORVF
4-
/RAW VIDEO IMAQE /
PPPSSMMY.RVF /
1 4.
/
/ uvA/i HATA /
EMAPPOLY.RVF /
•1
RESAMPLE
4
/WARPED IMAGE /
PPPPWARPflVF /
4,
MOSAIC
4
/ "TA /
\MANUAL /
OPERATION /
•WCES,
NAMED
MODULE
1 STOBEDDATA (
1 CONNECT ]
/MOSAICED IMAQES /
MPPPMMYYJWF /
Jr
/
/
/
/FEATURE MAP /
OUTPUT HA3TEI1 M
PPPMMYYFJtVP /
1
VECTORIZE
i
DATABASE REPORT
1
/CENTROID TEXT /
RLE /
CFPPPMMYY.TXT /
1
/CENTROID /
SASDATA /
CFPPPMMYY.SSD /
\ FFATUf
IE MAP /
/ SAVED FEATURE
/ UAP
/
Ny^ INTERPRETATION / J PPPMMYYJIVF /
.
\FEATUF
TRANS
^
r
IEMAP /
/ NWIDATA
LABEL f^ 1 EMAPPOLY.HVF >
r
/FEATURE MAP /
OUTPUT TEXT FILE /
DPPPMMYY.TXT /
T
r
CONVERT TO SASDATA
i
/FEATURE MAP /
OUTPUraASFILE /
DPPPMMYY.3SD /
SW>
i ,
0
2
/
Vpp««Jx4-2tord«an«afproc«li»».
Figure 4-2. Process for creating digital files from video tape.
29
-------
- 9- "" * 1
-•*-. f
Figure 4-3. Example of final raster from video data for plot 374 after georeferencing and resampling.
These video scenes were the base data used for interpretation of the presence of water. The process
was accomplished by the feature mapping procedure in MIPS. Two tasks were accomplished during
feature mapping, interpretation of areas as water and transfer of attribute data from the vector files
(translabeling process). At this stage we saved a raster showing the area interpreted as water. We also
saved a featuremap output raster that was used for additional processing (Fig. 4-4).
30
-------
**.•*
% *
. •* »
V "
-------
4.2.3 Aerial Photographs
We took aerial photographs of all 10.4-km2 plots and sample wetland basins in June of 1992
and 1993 on the dates shown in Table 4-2. The camera was a Nikon F2 35mm equipped with a Nikon
50mm lens and a Tiffen UV filter.
Table 4-2. Dates on which photographs were obtained during a pilot study of indicators of wetland
condition in a pilot study of indicators of wetland condition in the Prairie Pothole Region
of the United States."
Dates
Wetland Basin
1992
1993
038
054
059
060
073
133
134
145
156
241
246
249
327
363
374
396
407
441
442
498
06/18
06/20
06/20
06/20
06/18
06/20
06/20
06/18 06/23
06/26
07/18
06/26
06/18
06/18 06/26
06/26
06/23
06/23
07/14
07/14
06/18
06/15
06/15
06/15
06/18
06/10
06/10
06/10
06/10
06/15
06/15
06/10
07/20
07/20
07/20
07/20
"For some plots, weather conditions required photography of parts of the plot on different dates,
32
-------
Film was Ecktachrome 100, The camera was hand held and sighted through a belly port in a
Cessna 185. High-level photographs that covered entire 10.4 km2 plots were taken at 3,353 m AGL.
Low-level photographs of individual wetland basins and their surrounding uplands were taken at
elevations of 686 to 1,372 m AGL, depending on the size of the target wetland basin. The film was
processed by the Kodak E6 process and mounted to 35 mm slides. We used a Nikon LS 3500 SR1
slide scanner at 835 dots/cm to create MIPS *.RVF files from the slides. Processing of the rasters from
the slides to register the data, eliminate distortion, and mosaic into a single raster covering each 10.4
km2 plot was the same as that used for aerial video (Fig. 4-5). Data from both video and high-level
slides (Fig. 4-6) were used for photointerpretation of features that did not appear on the NWI data.
These new features were identified and processed to create updates to the vector data (Fig. 4-7). The
criteria for classifying these new figures are given in Table 4-3.
Table 4-3.
Wetland classes used for water areas that did not appear in NWI mapping.
Class
Criterion
New Wetland
Dugout
Stock Pond
Partially-drained
Wetland
Drained Wetland
Unclassified
"Drain
Natural Drain
Held water two or more months. No obvious berm present.
Held water two or more months. Obvious berm present.
Held water two or more months. Dam structure present across
drainage.
Held water at least one month. Had basin shape. Obvious drainage
channel from basin.
Did not show water, but had obvious basin shape. Obvious drainage
structure present.
Water present less than 1 month. No obvious outlet present.
Considered ephemeral. Does not meet wetland definition.
Situated between two or more wetland basins. Water visible at least
one month and clear line for at least two months. No obvious artificial
or enhanced natural drainage.
Obvious natural drainage way that does not appear in NWI data.
33
-------
HUH LEVEL
MM DATA /
EHMKH.rj»F /
DATA
LEVEL MAOES
(•AMMO* 4^4 tor 4Mb gfpmduM.
Figure 4-5. Process used to derive digital data from 35-mm slides.
We took low-altitude 35-mm photographs of each sample wetland basin in support of the
vegetation and soils studies. This process (Fig. 4-8) required making Cibachrome prints from the low-
level photographs. Zones of vegetation and the location of sample quadrats were delineated on the
photographs in the field. The prints were then scanned and georeferenced. Vectors were manually
drawn over the field delineation to obtain a data set showing the vegetation zones and location of the
sample quadrats.
34
-------
>
,
Figure 4-6. Example of raster from high-altitude 35-mm photograph after georeferencing and resampling.
4.2.4 Analyses
We did not conduct statistical tests of differences in cases where changes in condition would
have no effect on the landscape attributes. Instead, we present descriptive statistics (median,
dispersion in units of Hspread, and extreme values) as box plots (Velleman and Hoaglin 1981) because
of the extremely skewed distribution of the data and the frequent occurrence of outliers. The outliers
-r
-------
Figure 4-7. Process for adding vector data for features that do not appear in mapping by the National
Wetland Inventory.
were identified by plot number for reference to the appendices. These analyses do not include the
class, Lake.
36
-------
We developed an index designed to evaluate monthly changes in the amount of water in
wetland basins. The index was:
where Cl is the change index, w( is either the area of water or number of ponds in month 1, and Wnwi is
either the area of wetland or the number of basins from the NWI data. The devisor is used as a scaling
factor. We calculated Cl for each wetland basin class in each year of the study. The data underwent
analysis of variance (ANOVA) for a repeated measure split-plot design. The whole plot treatment was
condition (good and poor). The subplot treatment was wetland basin class (temporary, seasonal, or
semipermanent), and the repeated measure was year (1992 and 1993).
To test for differences between condition classes for number of drained wetlands, length of
drainage ditches and area of cropland, we used the TTEST procedure (SAS Inst. 1989). Because of the
unequal variances, we used Satterwhite's approximation for degrees for freedom (Steele and Torrie
1980).
4.2.5 Wetland Drainage Basins
Our original project plan called for the use of a Geographic Positioning System (GPS) for
delineating drainage basins for each sample wetland basin. We conducted tests of the precision of
elevation measurements derived from the GPS and found that they were not precise enough to
delineate drainage basins in our study areas that have little topographic relief. In addition, a field test
demonstrated that the method required more time than we had available. To obtain a crude
measurement of drainage basins for each sample wetland, we used four field measurements of the
distance from the edge of the wetland basin to the divide between basins. We then interpolated the line
between the four points by referring to aerial photographs, topographic maps, and field notes. In those
cases where only a portion of the entire wetland basin was used as a sample site, we truncated the
drainage basin where it extended beyond the area from which we obtained data. Drainage basin
delineations were digitized and intersected with the vector data for the plot as shown in Fig. 4-9.
37
-------
UHT
, ,„, /
MiaVHElATIOH /
•CMTOHVF
/wnrnoro /
•1MB /
mwmajnp /
7
CHEMTE VECTOR
/VECICRFEK /
reCTtwnfjWF /
MTA
tmataiaA
r^-i
I jBgrFm /
/ VPPPPtC.TXT /
1
OONVamOSASMTA
Figure 4-8. Process used for creating SAS data sets from low-elevation photographs and field delineation.
4.2.6 Duck Populations and Production
Our analysis of duck populations and production were derived from model predictions. Duck
count data were supplied by the U.S. Fish and Wildlife Service (USFWS) and were used in conjunction
with the pond estimates and upland cover availability determined during this study. No actual counts of
ducks were made on the sample wetland basins used in this study. Our estimates are model
38
-------
-projections. Cowardin et al. (1995) described the methods used by the USFWS and in this study.
Breeding pair estimates from the size of individual ponds were the result of a regression model
(Cowardin et al. 1988) that predicts breeding pairs from the sizes of individual ponds present in May.
Breeding pair estimates do include the class Lake, which was excluded during this study. The pond
estimates were from video taken during the first week of May in 1992 and 1993 (see Section 4.2.2).
These estimates were corrected for regional and annual variation by using estimates of 7 (total number
of counted pairs/number of pairs predicted by a regression model, see Cowardin et al. 1995) for the
Wetland Management District containing our 10.4 km2 plots. Recruitment estimates were derived from
habitat availability estimates for our 10.4 km2 plots and nest survival estimates from Shaffer and
Newtown (1995). Production of mallard recruits was a model prediction based on the product of
breeding population and recruitment estimates for each plot derived from the model of Johnson et al.
(1987).
We suspected that variation in duck counts was primarily due to differences in numbers of
wetland basins on each plot. To test this, we used a repeated measures analysis of covariance where
the covariate was the number of basins. This analysis indicated that the number of pairs did not depend
on the number of basins (F212 = 3.38, P = 0.68). Therefore we used a repeated measures analysis of
variance where years was the repeated measure and the main effects were species and condition. The
response variable was transformed by the 1n(y+1) transformation and the analysis was conducted by
the general linear models procedure of SAS (SAS Inst. 1989). The analysis method used to test for
difference between condition classes and years was the same as that used for pairs.
4.3 RESULTS
4.3.1 Wetland Abundance and Distribution on Sample Areas
The abundance and distribution of wetlands on the sample 10.4 km2 plots, with the exception of
drainage and construction of wetlands, is not the result of human-induced changes. Rather it is a
characteristic of the geologic setting of the plots and should not be considered as indicating condition of
the landscape. We present these data on basin wetland abundance and distribution because they tend
to confound some of the analysis of indicators of condition that follow. The tremendous variation
exhibited in the data is also important for planning future probability sampling.
39
-------
Figure 4-9. Process used for creating SAS data sets representing drainage basins of each sample wetland
basin.
The number of wetland basin/pilots was highly variable and the distribution was skewed. Four
plots had densities greater than 250, and there were two plots with more than 300 wetland basins
(Fig. 4-10). Although the median wetland density was similar for good-condition (Median [x] = 15) and
poor-condition (x = 18), the poor-condition plots were more variable.
40
-------
400
,_ 300
3
Q^
|200
%
2 100
UJ
m
D
z 0
-100
O 374
-
™
O 363
* 156
T
1.. I
— 3- — 1
i —
0 134
-
-p -
—
GOOD POOR
CONDITION
Figure 4-10. Distribution of wetland basin density for plots classified as in good- and poor-condition.
The area of wetland/plot also showed similar medians between good-condition (x = 11.3 ha)
and poor-condition (x"= 8.4 ha) plots, and a distribution skewed to the smaller areas in both condition
classes. Good-condition plots showed more variation except for three outliers (plots 134, 135, and 442)
among the poor-condition plots (Fig. 4-11).
The mean distance of each basin to its nearest neighbor (Fig. 4-12) was similar between good-
condition (x = 387.9 m) and poor-condition (x = 268.2 m) plots except for plot 241 in the poor-condition
group. This plot had only three wetland basins and these were widespread.
The shoreline development index (SDI) (Cole 1983) compares the boundary of each wetland
polygon to that of a circle with the same area. The index has a value of 1 for a perfect circle. Median
values of SDI for good-condition (x =1.4) and poor-condition (x =1.3) wetland basins (Fig. 4-13) were
similar and there was little variation except for plot 241, where there were a number of streams that
have high indices because of their long, narrow shape.
41
-------
80
5 60
Q 40
£ 20
u_
O
UJ
oc o
-20
134
442
145
GOOD POOR
CONDITION
Figure 4-11. Distribution of area of wetland/plot for good- and poor-condition plots.
4.3.2 Drainage as An Indicator of Wetland Landscape Condition
The characteristics discussed thus far are not necessarily related to condition of the landscape
of the plots; rather they are characteristics of the geomorphology of the setting where the plots exist.
Drainage of wetlands is probably the most extreme factor affecting wetland condition because, once
drained, the basin loses all wetland functions and their associated values. Although some drainage did
occur on good-condition plots, the number of drained wetland basins (good-condition x = 1, poor-
condition x = 12.5) was higher and more variable on poor-condition plots (Fig. 4-14). The presence of
drainage ditches is an additional indicator of condition of wetlands in the landscape. In many cases
ditches may be present from wetland basins that have not been completely drained. These basins
remain as wetlands but their hydrologic function has been severely modified by the ditching. The length
of drainage ditch per plot (Fig. 4-15) was greater and more variable on poor-condition plots (x = 11.2
km) than on good-condition plots (x = 3.2).
42
-------
3000
1
z
3
a 2000
Ul
1
8
wp
s
I
1000
O 241
GOOD POOR
CONDITION
Figure 4-12. Distribution of mean distance from wetland basins to nearest basin between good- and poor-
condition plots.
4.3.3 Seasonal and Annual Change in Ponds
The number of ponds and the area of wetland covered by water changed within and between
years. These estimates were confounded as an index to wetland condition because both the number of
wetland basins and the area of water depend on the geomorphology where the plots are located and
geomorphology cannot be considered an indicator of wetland condition. Therefore we evaluated
estimates of the percent of wetland basins and the area of wetland containing water in 1992 (Fig. 4-16)
and 1993 (Fig. 4-17).
Temporary basins were less variable. Poor-condition semipermanent basins had a consistently
smaller percent of the wetland area covered by water in both years. This would be expected because
the size of the poor-condition semipermanent basins (x = 2.2 ha) is much smaller than the good-
condition semipermanent wetland basins (x = 23.8 ha) and within the class semipermanent water in
large ponds is more permanent than in small ones.
43
-------
6
2
0
i4
Ul
•g
s
uj 3
5
D
Ul
^p
I2
CO
1
i i
O 241
T
I — "" — I
I 1
GOOD POOR
CONDITION
Figure 4-13. Distribution of shoreline development indices between good- and poor-condition plots.
Results for indices to change in area covered by water were similar to those for numbers of
ponds (Fig. 4-18). The index to wetland change in area of water for the June-July period was
significantly different between condition classes (F, 14 = 7.71, P = .015). The least squares mean for
poor-condition was 0.231 and for good-condition 0.051. There was also a significant year effect for the
May-June Period (F141, = 4.35, P = 0.043).
Indices to change in pond numbers (Fig. 4-19) differed between good-condition and poor-
condition plots for all wetland basin classes in the June-July interval (Ft 14 = 15.81, P = 0.001). Poor-
condition plots had a least squares mean of 0.287 and good-condition plots had a least squares mean
of 0.062. Indices also differed between condition classes for semipermanent ponds in the May-June
(F227 = 4.16, P = 0.03) period and for the mean of all periods (F227 = 3.84, P = 0.034). There was a
significant (F, 41 = 5.81, P = 0.02) difference in change in pond numbers between years. Least squares
means were 0.224 (SE = 0.035) for 1992 and 0.125 (SE = 0.035) for 1993. No differences were
detected for index to change in pond numbers for the April to May period.
44
-------
40
3
0.
z 20
§
CD
Q
W 10
<
cc
Q
-10
I
1
1
GOOD POOR
CONDITION
Figure 4-14. Distribution of drained wetland basins on good- and poor-condition plots.
4.3.4 Upland Characteristics of Study Sites
Our selection procedure was designed to maximize the difference in cropland/upland ratio and
as expected cropland was dominant on poor-condition plots (x"= 951.1 ha, Fig. 4-20). Conversely
grassland dominated good-condition plots (x* = 420.7 ha). Cropland was still an important component of
the good-condition plots (x*= 248.5 ha), whereas grassland was largely absent on poor-condition plots
(x"= 0.0). There was no hayland, planted cover, scrubland, or woodland on poor-condition plots. The
remainder of the upland cover classes were similar between good- and poor-condition plots, except for
CRP cover which was more abundant on good-condition plots. There was a significant difference (t75 =
5.999, P = 0.0007) in the area of cropland between good- and poor-condition plots. The good-condition
45
-------
E 40
I
D.
5
UJ
S
D
u_
O
1
ui
30
20
10
0
•10
* 54
T
GOOD POOR
CONDITION
Figure 4-15. Distribution of lengths of drainage ditch per plot for good- and poor-condition plots.
plots were much more variable (x = 297.62, SE = 110.75) in the amount of area of cropland present
than the poor-condition plots (jf = 027.84, SE = 20.05). The distribution of cropland among the poor-
condition plots was narrow with a median value of 951 ha which represented about 92% of the plot
area (Fig. 4-21). The outlier plot (442) contained 20.9 ha of CRP cover.
Our analysis of the amount of exposed soil subject to erosion in June showed a significant
difference between condition classes (t73 = 3.0254, P = 0.0184). Good-condition plots had a mean of
67.54 ha (SE = 44.903) of exposed soil; poor-condition plots had a mean of 983.19 ha (SE = 299.304).
The medians were 629 ha for poor-condition plots and 0.0 for good-condition plots (Fig. 4-22). The
poor-condition plots were also more variable than the good-condition plots. Poor-condition plots had two
outliers with more than 2000 ha of exposed soil subject to erosion.
46
-------
cc
UJ
I
CD
UJ
8
UJ
£L
TEMPORARY
Jura
July April
SEASONAL
so.
40.
ao.
10.
so
40
30
10
April
July April
SEMIPERMANENT
100
BO.
20.
Jura
July
PONDS
GOOD CONDITION -
April May Jura
AREA OF WETLAND
POOR CONDITION —
July
Figure 4-16. Percent of wetland basins containing water and percent of wetland area covered by water for
each wetland basin class in 1992.
47
-------
cc
UJ
I
8
H
Ul
HI
0.
TEMPORARY
25
20
IS
10
\
\
\
\
\ :
April
JUM
July
SEASONAL
90
50
40
30
an-
10
June
July April
SEMIPERMANENT
May
Jum
July
SO
40
20
10
100'
00
40
April
JUM
PONDS
GOOD CONDITION -
JUM July April May
AREA OF WETLAND
— POOR CONDITION
July
Figure 4-17. Percent of wetland basins containing water and percent of wetland covered by water for each
wetland basin class in 1993.
48
-------
TEMPORARY
SEASONAL
cc
I
u.
O
s
s
z
111
O
D
Ul
e
SEMIPERMANENT
0.1
0.1
GOOD CONDITION
POOR CONDITION
APR-MAY MAY-JUNE JUNE-JULY MEAN
APR-MAY MAY-JUNE JUNE-JULY MEAN
APR-MAY MAY-JUNE JUNEJULY MEAN
Figure 4-18. Comparison of indices to change in pond between months by wetland class and period.
Asterisk indicates (P<0.05).
49
-------
4.3.5 Duck Populations and Production
Our analysis showed that there was a significant year by condition interaction (F170 = 4.17, P =
0.0448); therefore, we could not interpret main effects, year and condition. Differences in
condition-within-year and year-within-condition class were tested using Fisher's least significant
difference procedure (Milliken and Johnson 1984). There were more ducks predicted on good-condition
plots than on poor-condition plots in both years (1992, P = 0.0038; 1993 P < .001) (Fig. 4-23). There
was no significant difference between years within condition class in both condition classes ( P =
0.0628 for good-condition and 0.3212 for poor-condition). There was a significant species effect (F45e =
21.21, P <0.001). The individual species effects were not of interest and no tests were conducted for
comparisons among the five species. Analysis of mallard recruits showed that mallard recruits did not
vary significantly with number of wetland basins (F212 = 2.90, P = 0.0939). No differences were
detected between years (F, 14 = 2.87, P = 0.1124) or condition classes (F1i14 = 1.45, P = 0.2492). The
condition class-by-year interaction was not significant (F, 14 = 1.16, P = 0.2994). However, the means
were higher for good-condition plots (27.28 recruits/plot) than for poor-condition plots (14.81).
4.4 EVALUATION AND RECOMMENDATIONS
Results from this study show that remote sensing of physical parameters of selected
landscapes can furnish data to evaluate the condition of those landscapes. Though the ratio of cropland
to upland was used as a proxy for wetland condition, we believe that the direct measurement of the
amount of cropland in a landscape is probably the simplest and most meaningful indicator of condition
of wetlands in that landscape and that it is easily obtained from base mapping that does not need to be
conducted annually. Although the amount of cropland in an area is relatively constant, there can be
major changes resulting from agricultural programs such as the CRP program, which converted large
amounts of cropland to grass/legume cover. We obtained our estimates from baseline mapping
conducted by NWI, but we suspect that satellite data with resolution equivalent to LANDSAT (30 m
pixel) could be used to monitor the amount of cropland in the PPR. The main advantage would be that
coverage of the entire area is possible. The only constraint would be the cost both for data and for
processing.
We recommend that all wetlands and uplands be mapped and that digital data sets be
prepared from the maps prior to initiation of planned work on sample 40-kmz hexagons. Such data
would not only furnish a measure of landscape condition at the time of mapping, but would also be
50
-------
TEMPORARY
SEASONAL
SEMIPERMANENT
0.5-
GOOD CONDITION
POOR CONDITION
APR-MAY MAY-JUNE JUNE-JULY
APR-MAY MAY-JUNE JUNE-JULY
APR-MAY MAY-JUNE JUNE-JULY MEAN
Figure 4-19. Comparison of indices to change in area covered by water by wetland basin class and period.
Asterisk indicates (P<0.05).
51
-------
1000-,
Legend
GOOD CONDITION
POOR CONDITION
BARREN
CROPLAND
HAYLAND PLANTED COVER SCRUBLAND
WOODLAND
Figure 4-20. Area of upland habitat classes/plot by condition class.
essential to registering other remote sensing data to a common map projection so that GIS analyses
may be conducted. Once the baseline data have been collected, the data sets can be easily be
updated by remote sensing and ground survey methods, thus producing a temporal series of GIS
layers.
4.4.1 Drainage as An Indicator of Condition
We used two indicators that are direct measures of wetland condition, number of drained
basins and length of drainage ditches. These indicators effectively separated good- and poor-condition
plots according to our definition. These indicators were correlated with the amount of cropland (drained
basins, r2 = 0.56, drainage ditches r2 = 0.57) but there are often situations where the amount of
drainage may be different and unrelated to the amount of cropland present. The differences may be
related to geomorphology of the area and to various Federal or State programs that encourage or
52
-------
constrain drainage. For example, an appreciable portion of the wetlands in the PPR are under
perpetual USFWS easements that prevent drainage.
Our estimates of drainage were based on mapping by NWI and on interpretation of aerial vedio.
These methods relying on interpretation are subject to both variation caused by the interpreter and to
bias, primarily where the interpreter commits errors of omission. We suspect that our estimates of the
length of drainage ditch are conservative because it is often difficult or impossible to see tile drainage
on an aerial photograph. Tests of these error were beyond the resources of the pilot. Measurement of
drainage requires more resolution than measurement of the amount of cropland. The 1:63,000
photographs used by NWI in combination with our low level video and photography were adequate. We
stress that our estimates are of drainage that has taken place over a long period of time. It would be
AREA OF CROPLAND (ha)/PLOT
-
-p
I ' [
* 442
GOOD POOR
CONDITION
Figure 4-21. Distribution of cropland/plot between good- and poor-condition plots
53
-------
advantageous that EMAP have the ability to monitor annual change in these two indicators, even
though data need not be gathered each year. The annual rate of drainage is low and drainage is often
clumped (all wetlands may be drained on some areas where others remain undrained). This
characteristic of drainage means that large samples will be required to obtain precision in annual
estimates.
We recommend that drainage of wetland basins and creation of drainage ditches be monitored
for each sample 40-km2 hexagon. The repeat schedule of 4 years in the EMAP sampling protocol
(Leibowitz et al. 1991) would be adequate to detect long-term changes in loss of hydrologic function
due to drainage.
3000
2000
I
EXPOSED SOIL
i
-
0248
A 306
GOOD
**£
-
i
POOR
CONDITION
54
-------
ttO
1992
GOOD CONDfTION
POOR CONDITION
MAUAflD OWMU.L B-WTEAL SHOVELER PIN1WL
1093
TOTAL
GOOD CONDITION
POOH CONDFTION
IMLLARD
QMMIHU
B-WTEAL
•HOVELBI
PMTAL
TOTAL
Figure 4-23. Numbers of five species of ducks per 40-kmz hexagon on good- and poor-condition plots in
1992 and 1993. Asterisk indicates significance at 0.05 level.
55
-------
4.4.2 Area of Exposed Soil
Our low level 35-mm photographs allowed us to estimate the area of exposed soil on each plot
in June. This indicator effectively separated good- and poor-condition plots. The area of exposed soil is
important because the amount of silt and probably contaminants moving from the uplands into the
wetlands is not just the result of the fact that an area is in cropland but also of what is growing on that
cropland. There is also reason to believe that material moving to the wetland is a function of the type of
crop and its stage of development. Unfortunately, we were not able to interpret crop type and
development stage from our photography. To do this we would need methods that improve both spatial
and spectral resolution.
We recommend that this estimate of the area of exposed soil as an indicator of landscape
condition be continued. Furthermore, we recommend that research be conducted to find technology that
furnishes better spatial and spectral resolution. This enhanced capability would assist in documenting
actual crop types present within the hexagons. Without data on crop types, interpretation of results from
monitoring biological indicators of condition would be difficult, and direct measurement of herbicide and
pesticide contamination also would be more difficult. For example, results presented in Chapter 8 show
that the use of atrazine is correlated with the presence of corn.
4.4.3 Index to Wetland Change
Our index to wetland change was able to separate good- and poor-condition landscapes. We
suspect that agricultural tillage of the wetlands and the surrounding uplands has altered the hydrologic
function of the wetlands. Wetlands in disturbed sites are apparently hydrologically less stable than
those in grasslands. The index to change in pond numbers was more sensitive than the index to
wetland area. The change between June and July was the best separator of good- and poor-condition
landscapes.
We recommend that this index be calculated for 40-km2 hexagons during the next phase of
EMAP in the PPR. However, our data suggest that the change from April to May does not furnish a
good indicator of condition. Water in wetland basins may be frozen during April and snowcover was
often present and obscured the basins; therefore, interpretation of water was inaccurate and the
estimate of change from April to May is unreliable. We recommend that no April flight be conducted in
the next phase of EMAP.
56
-------
4.4.4 Estimates of Duck Production
Our model-based estimates of breeding population and of mallard production detected
differences between good- and poor-condition landscapes. The difference was apparent despite the fact
that we observed no association between duck pairs and number of wetland basins as would be
expected. Number of breeding pairs showed promise as an indicator of landscape condition, but there
were two problems. The counts were extremely variable and our sample sizes were meager. In
addition, the model derived breeding pair estimates depend on ground counts that were made in the
Wetland Management District, but not on the sample 10.4-km2 plots. The resulting correction (y) had
outliers because of atypical wetland basins that were not actually on the plots. For example y for
mallards was 18.9 in 1992 and 14.0 in 1993 for the Crosby-Lostwood Wetland Management District
which contains poor-condition plots 441 and 442. These extremely high y estimates result from a single
wetland basin in the district but not on the 10.4-km2 plots.
We recommend that estimates of the five dabbling duck species used in the pilot study be
continued, but that pair counts used to estimate y be made on the 40-km2 hexagon in the next phase of
the EMAP studies. Three factors will greatly improve usefulness of this indicator of condition: (1) The
larger size of the sampling units (40-km2 hexagon) will help to reduce variability among sample plots;
(2) the larger sample of plots (45 versus 16 in the current pilot) will improve chances of detecting
differences between condition classes; (3) if we conduct the pair counts on the sample 40-km2
hexagons, the procedures used in the pilot will have more validity. The procedure also will help solve
the problem of obtaining a valid sample for each hexagon. We do not have to assume that basins
where ducks are counted represent population density for the entire polygon; we only need to assume
that deviation from our regression estimate for all ponds is represented by our sample. In practice, we
recommend that the sample of ponds for estimating y be a roadside sample which represents the size
classes present on the hexagon. The actual estimate of duck numbers will be derived from video of the
entire hexagon and can be corrected for temporal and spatial differences by y derived from the
roadside sample, thus avoiding the land access problem.
57
-------
Section 5.0
IMPACT OF AGRICULTURAL PRACTICES ON
WETLAND MACROINVERTEBRATES,
SILTATION RATES, AND WATER-LEVEL FLUCTUATIONS
Ned H. Euliss, Jr. and David M. Mushet
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
5.1 INTRODUCTION
In support of the EMAP goal of collecting time-integrated measures of wetland condition, we
evaluated the use of aquatic invertebrate remains, siltation rates, and water-level fluctuations in
wetlands as indicators of wetland condition. Because the initial development and testing of sampling
devices began in 1992 (see Section 9), they were not used on the EMAP pilot study plots until 1993.
Hence, this section is based on a single year's sampling. We had originally planned to include water
quality, in-situ invertebrate, and amphibian measurements in this study, but drought conditions in 1992
precluded their use and they were officially dropped from consideration.
5.2 OBJECTIVES
1. Determine if aquatic macroinvertebrate recalcitrant remains can be used to distinguish
between wetlands occurring in poor-condition (cropped) landscapes and good-condition
(grassland) landscapes.
2. Determine if siltation rates can be used to distinguish between wetlands occurring in
poor-condition landscapes and good-condition landscapes.
3. Determine if water-level fluctuations can be used to distinguish between wetlands
occurring in poor-condition landscapes and good condition landscapes.
59
-------
5.3 METHODS
5.3.1 Objective 1
To determine whether aquatic macroinvertebrate remains could be used to distinguish between
good-condition and poor-condition landscapes, we collected sediment and macroinvertebrate samples.
From October 14 to October 25 (before freeze-up) and from April 21 to May 13 (after freeze-up) in
1993, we installed 5 bottle-top sediment traps (see Section 9) in each of the 36 EMAP pilot study
wetland basins. Sediment traps served to collect both macroinvertebrate remains as well as sediments
entering wetland basins from surrounding land. Within each wetland basin, we installed one trap on
each of five transects that radiated from the center of the wetland basin (defined here as the lowest
elevation) along random compass bearings.
We installed the sediment traps so that the top of the traps were 7.3 cm above the
sediment-water interface and at an elevation where the tops of the traps would be level with the water
surface when water depth at the wetland's center was 10 cm (Fig. 5-1). We used a Spectra-Physics
Model 650 Laserplane to determine all elevations within +1.6 mm per 30 m. Sediment traps installed in
wetland basins grazed by livestock were covered with a steel tripod surrounded by a length of chain to
reduce disturbance.
At each study wetland, we located a large, stationary object (tree, power pole, large boulder,
etc.) and marked it with high-visibility paint to serve as a benchmark to evaluate the effects of frost
upheaval on sediment traps. We determined reference benchmark elevations (+1.6 mm per 30 m) using
a laser level and then measured and recorded the difference in elevation between the benchmark and
the tops of the sediment traps. We measured this difference in elevation again in early mid-spring after
the wetlands had become ice-free to determine if freezing upheaval had altered the positions of the
sediment traps. We readjusted the elevations of the sediment traps as necessary.
Just before fall freeze-up in September 1993, we removed all the sediment traps from wetlands
and transported them back to the Northern Prairie Science Center (NPSC) laboratory in Jamestown,
ND, where samples were stored in freezers until processed. We processed samples by removing a
sample from the collection tube while it was still frozen, concentrating residues from the thawed sample
on a 0.5 mm screen, examining sample residues over a light table, and separating invertebrate
recalcitrant remains from residues using forceps. Soil and other debris > 0.5 mm remaining in the
60
-------
7.3[cm
J2.
cm
T
Wetland
Center
Figure 5-1. Placement of sediment traps in EMAP pilot study wetlands.
sample residue was returned to the screened sediment sample for determination of sediment dry
weights. We then sorted the invertebrate remains into major taxonomic groupings, and enumerated and
weighed them to the nearest 0.0001 g on an analytical balance after drying to a constant weight at
55-60 °C.
Statistical Methods: The response variables we analyzed included taxon richness, biomass,
and abundance. The response variable for the biomass and abundance analyses was the mean weight
(g) or count for each taxon (or all taxa combined). Taxon richness was the total number of taxa
observed in each wetland basin. Separate analyses were performed for each of these
macroinvertebrate response variables using the SAS General Linear Models (GLM) procedure (SAS
Inst. Inc. 1989). Two-way analyses of variance (ANOVA) were used to assess the effects of wetland
class (temporary, seasonal, and semipermanent), condition (good and poor), and their interaction on
response variables. Rsher's least significant differences (LSD) procedure (Milliken and Johnson 1984)
61
-------
was performed to assess significant differences among wetland classes. We included only those
wetlands in the sample that contained water some time during 1993 in our analyses (32 wetlands).
Taxon richness data consisted mostly of small whole numbers with numerous zeros. A
transformation was performed using the square root of taxon richness plus 0.5 to stabilize the variance
(Steel and Torrie 1980). A logarithmic transformation of the invertebrate count plus 1.0 was used to
stabilize the variance to facilitate the abundance analyses (Steele and Torrie 1980). This transformation
was also applied to the biomass data; however, the results of the biomass analysis in this report are for
the untransformed data because both yielded similar results. For the analysis of the abundance data,
we pooled over all transects and taxa within each wetland basin. In addition, separate analyses were
performed for the four most common taxa (Cladocera, Ostracoda, Planorbidae, and Lymnaeidae).
5.3.2 Objective 2
To test siltation rates as an indicator of wetland condition, we separated sediments from
invertebrates collected in our sediment traps by screening and removing invertebrate remains by hand.
A 0.5 mm mesh screen retained invertebrates and larger sediment particles and debris, while fine
sediments passed through. After we removed invertebrates from sample residues, we added the
remaining material to the fine sediment material that was previously separated by sieving. We then
weighed all sediments collected in the traps described to the nearest 0.01 gram after we centrifuged
them at 5,000 rpm for 10 minutes to remove excess water and dried them in an oven at 100 °C until a
constant weight was reached.
Statistical Methods: The analysis was performed using the SAS GLM (SAS Inst. Inc. 1989). A
two-way ANOVA was used to test for condition effects, class effects, and the condition by class
interaction. If the class effect was significant, a Fisher's LSD procedure (Milliken and Johnson 1984)
was used to isolate the location of differences. Only wetlands that contained water during 1993 were
used in the analysis (n=32 wetlands).
In the analysis, we averaged sediment dry weights over all transects within each wetland. This
was necessary because of the sparse nature of the data. The logarithmic transformation of the
response plus 1.0 was used to stabilize the variance to facilitate the analysis (Steele and Torrie 1980).
62
-------
5.3.3 Objectives
We used a water-level recorder to find out whether water level fluctuations could distinguish
between wetlands occurring in good- and poor-condition landscapes. From April 21 to May 13, 1993,
we placed one water-level recorder in the center (lowest elevation) of each of the 36 EMAP pilot study
wetlands. The water-level recorders were developed specifically for this EMAP pilot study and are
described in Section 9 of this report. Briefly, they consist of a copper-coated steel rod that guided a
large float up and down as water levels fluctuated. Two magnetic slides, one above and one below the
float were pushed by the float to positions on the rod that corresponded to maximum and minimum pool
levels. The devices were removed from the wetlands between Aug. 25 and Sept. 16,1993, and the
distance between the 2 slides was measured and recorded. The distance between the slides was our
measurement of water-level fluctuation during the study.
Statistical Methods: The analysis was implemented by SAS GLM (SAS Inst. Inc. 1989). Two-
way ANOVA techniques were used to assess the effects of condition (good and poor), class
(semipermanent, temporary, and seasonal), and the condition by class interaction. If the class effect
was significant, a Fisher's LSD test (Milliken and Johnson 1984) was used to locate differences. The
response evaluated was the difference between the maximum and minimum depth measurement
divided by the total area of the watershed (see Section 7). A logarithmic transformation of the response
plus 1.0 was used in the analysis of response variables to stabilize the variance to facilitate the analysis
(Steele and Torrie 1980). Only wetland basins that contained water during the study and basins where
the devices were not destroyed by cattle were used in the analysis (n=27 wetland basins).
5.4 RESULTS
5.4.1 Objective 1
Our analyses of taxon richness, invertebrate biomass, and invertebrate abundance
suggest that we may have collected too few samples (n=5) from an insufficient number of wetland
basins (n=27) (Table 5-1). Based on the variance observed in our taxon richness data, we estimate that
we should have collected 3 to 11 samples (16 to 148 for biomass) from 100 to 120 wetland basins (440
to 460 for biomass) just to estimate within 10% of the mean, 90% of the time. Based on our small
sample size of 32, we failed to detect differences in taxon richness for wetland condition (F=0.01; 1,26
df; £=0.9111), wetland class (F=1.49; 2,26 df; £=0.2451), or for the condition by class interaction
63
-------
(F=1.28; 2,26 df; £=0.2952). However, there was a slight tendency for taxon richness to increase with
water permanence (i.e., temporary wetland < seasonal wetland < semipermanent wetland) (Fig. 5-2).
We also failed to detect differences in biomass for wetland condition (F=0.70; 1,26 df; £=0.4088),
wetland class (F=0.58; 2,26 df; £=0.5659), or in the condition by class interaction (F=0.50; 2,26 df;
£=0.6105).
Our analysis of invertebrate abundance also suggests that we collected too few samples from
an insufficient number of wetland basins, but the analysis did suggest that abundance may be worth
considering in future studies assessing water permanence. For all taxa pooled over the 5 transects,
there was a significant class effect (F=3.90; 2,26 df; £=0.0331) with semipermanent wetlands having
larger abundances (40.76) than temporary wetlands (3.45). However, we did not find a significant
condition effect (F=0.63; 1,26 df; £=0.4330) or a condition by class interaction (F=2.21; 2,26 df;
£=0.1297) for invertebrate abundance. We estimate that we should have collected 5 to 13 samples
from 650 to 670 wetlands to adequately evaluate the potential of invertebrate abundance as a potential
indicator of wetland condition (Table 5-1).
Table 5-1. Number of wetlands and samples per wetland needed to estimate means within 10%, 90%
of the time. Note that the number of samples per wetland depends on the number of
wetlands sampled.
Variance
Within Between
Variable Mean basin basins
Taxon Richness 1.3968 0.0829 0.1069
Biomass(g) 0.0094 0.0012 o.oooi
Abundance 2.7271 0.7192 2.7830
Sedimentation(g) 1.1783 0.2261 0.3866
Sample Size
Wetlands (Samples)
100 (11)
120 (3)
440 (148)
460 (16)
650 (13)
670 (5)
220 (13)
240 (4)
We also examined the effect of wetland condition and class on the abundance of the four most
common invertebrates in our sample (i.e., Cladocerans, Ostracods, Lymnaeid snails, and Planorbid
snails). For Cladocerans, we found no wetland condition effect (F=0.01; 1,26 df; £=0.9427), no wetland
class effect (F=1.57; 2,26 df; £=0.2263), and no interaction of condition with class (F=2.34; 2,26 df;
64
-------
^^^^H
o
E
i
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
Temporary Seasonal Semipermanent
Wetland Class
Figure 5-2. Taxon richness (back transformed LSMs) of invertebrates captured in sediment traps installed
in EMAP pilot study wetlands, 1993. Black bars = 95% C.I.
£=0.1165). Ostracods had no condition effect (F=0.31; 1,26 df; £=0.5837), a marginally significant class
effect (F=3.14; 2,26 df; £=0.0601), and no condition by class interaction (F=1.74; 2,26 df; £=0.1951).
Mean abundance of Ostracods in semipermanent wetlands (33.23) appeared to be higher than in
temporary wetlands (2.37). Abundance of Lymnaeid snails did not appear to be affected by wetland
condition (F=0.99; 1,26 df; P=0.3285), wetland class (F=0.23; 2,26 df; £=0.7985), or by the condition by
class interaction (F=0.74; 2,26 df; £=0.4853). However, Planorbid snails appeared to have been
affected by wetland class (F=2.88; 2,26 df; £=0.0743) with seasonal wetlands having more snails (1.31)
than temporary wetlands (0.08). Like the other four taxa of invertebrates, Planorbid snails were not
65
-------
affected by wetland condition (F=0.11; 1,26 df; £=0.7467) or by condition by class interactions (F=0.64;
2,26 df; £=0.5373). However, as with other invertebrate variables, sample size likely affected our
evaluation.
Our evaluation of the effect of frost upheaval on our sediment traps clearly indicates that
freezing winter temperatures did affect the locations of sediment traps (Table 5-2). Traps moved from 0
to 5.73 cm and had to be readjusted to the proper sampling elevation to insure that all sediment traps
collected invertebrates and sediments over the same time frame, regardless of water-level fluctuations.
Table 5-2. Mean movement (fall 1992 elevation - spring 1993 elevation) of sediment traps installed in
EMAP pilot study wetland basins during 1992. Plots 241, 246, 249, and 396 were dropped
as EMAP study sites in 1993. However, elevations were measured when we removed
equipment from basins within these plots in April, 1993.
Plot
Wetland f
Movement (cm)
Standard
Deviation
73
134
134
134
134
156
241
241
246
249
249
363
363
396
396
442
442
442
442
442
442
29
140
270
406
432
22
3
48
53
50
B6
22
56
106
107
93
260
261
281
295
301
0.00
0.18
2.32
-0.06
0.00
1.04
0.00
5.73
0.00
0.49
2.68
0.00
0.12
0.43
0.06
0.00
0.00
1.40
0.00
0.00
0.49
0.000
0.167
2.546
0.136
0.000
1.069
0.216
3.988
0.000
0.348
1.756
0.000
0.273
0.273
0.136
0.000
0.000
1.487
0.000
0.000
0.632
66
-------
5.4.2 Objective 2
The analysis of sediment dry weights, like our invertebrate analysis, failed to identify indicators
of wetland condition. Also, as with our invertebrate analysis, there was extremely high within-basin and
between-basin variability, and it appears that we collected too few samples from an insufficient number
of wetland basins. We found no wetland condition effect (F=0.18; 1,26 df; P=0.6775), no wetland class
effect (F=0.02; 2,26 df; P=0.9830), and no interaction of condition by class (F=0.80; 2,26 df; P=0.4618).
To have estimated the mean within 10%, 90% of the time, we estimate that from 4 to 13 samples would
have to be collected from 220 to 240 wetlands (Table 5-1); our small sample of 32 was clearly
insufficient to adequately evaluate this variable as an indicator of wetland condition.
5.4.3 Objectives
The analysis of water-level fluctuations (corrected for watershed size) clearly identified a
potential indicator for EMAP. We observed both a wetland condition effect (F=7.08; 1,26 df; P=0.0146)
and a wetland class effect (F=4.88; 2,26 df; P=0.0182). Further, there was no condition by class
interaction (F=0.86; 2,26 df; P=0.4376) to complicate the interpretation. In proportion to watershed size,
wetland basins in poor-condition had greater water-level fluctuations (14.14 cm) than basins in good
condition (4.27 cm). Further, seasonal wetlands and temporary wetlands both had greater water-level
fluctuations (11.82 cm and 13.74 cm respectively) than semipermanent wetlands (2.77 cm) (P=0.0220
and fM3.0090, respectively); there was no difference in water-level fluctuation between seasonal and
temporary wetlands (P=0.7775).
5.5 EVALUATION
5.5.1 Objective 1
Our analysis of the invertebrate data failed to identify invertebrate response variables that could
be used as indicators of wetland condition. While it is clear that we collected too few samples from an
insufficient number of wetlands, it is doubtful that even a sufficient number of samples would have
identified suitable indicators of wetland condition, with the possible exception of taxon richness.
Invertebrates are naturally highly variable, on both spatial and temporal scales, and this high natural
variability will make future attempts to identify invertebrate indicators difficult. We originally hoped that
67
-------
by using sediment traps to collect invertebrates over discrete time periods, much of this interfering
variability would be mitigated. However, our study clearly indicates that both the within-basin variability
and the between-basin variability was exceptionally high and that it would have required both an
inordinate and a cost-prohibitive number of samples to adequately evaluate the invertebrate response
variables.
Recommendations for Future Work: We recommend that invertebrate remains be dropped
from the next evaluation of wetland condition indicators. Of the response variables considered as
indicators, only taxon richness could possibly be evaluated, but we would need to more than double the
number of wetland basins sampled. If sediment traps are used to index taxon richness, our study
indicates that frost upheaval clearly alters the elevations of the traps and hence they will need to be
readjusted each spring after thaw. Despite the shortcomings of the approach used in this study, we still
feel that invertebrates have potential as indicators of wetland condition and suggest that future EMAP
work focus on invertebrate variables collected over a larger spatial scale (i.e., landscape scale). To this
end, the use of sticky traps or light traps (Belton and Kempster 1963, Harding et al. 1966, Belton and
Pucat 1967, Mason and Sublette 1971, Davidson et al. 1973, and Borror et al. 1981) may prove to be
particularly useful sampling methods.
5.5.2 Objective 2
Our analysis of sediment dry weights failed to identify an effective indicator of wetland condition
for EMAP. As was the case with the response variables evaluated under Objective 1, excessive
variation interfered with the evaluation of sediment dry weights as an indicator of wetland condition.
However, much of this variation may have been due to where the traps were placed in the wetland
basin rather than natural variability as has been documented for invertebrates.
Recommendations for Future Work: While it is intuitive that increased rates of sedimentation
would characterize landscapes heavily impacted by agriculture, we feel that the placement of our traps
strongly influenced the results. The sediment traps used in this study served the dual purpose of
collecting both sediment and invertebrate remains. As a safeguard against the generally dry conditions
during the initial year of the pilot, it was decided to place traps close to the center of the wetland basin
so they would collect invertebrates over longer periods of time. However, wetland basins tend to silt in
from the sides, and hence we probably did not place the sediment traps in an optimal location to
measure siltation. Future studies should use traps placed to measure siltation exclusively and locate
68
-------
them as close to the periphery of the basins as possible. Perhaps a better choice would be a surface
flow trap (Robert Gleason 1966) that is situated on the upland side of the wetland edge. The main
advantages of this trap over conventional sediment traps are that its function does not depend upon
water levels in the wetland basin and it is optimally placed to measure siltation near the wetland edge.
The main disadvantages are that vegetation growing in front of the collector tends to deflect silt-laden
runoff water and it is located close enough to the wetland basin edge that it could be damaged by farm
equipment, especially during dry years.
5.5.3 Objectives
The water-level recorders we developed (see Section 9) yielded data that clearly demonstrate
the value of water-level fluctuation as an indicator of wetland condition for EMAP. The water-level
recorder was also useful for separating wetland classes, and there was no wetland condition by class
interaction to complicate the interpretation of results.
Recommendations for Future Work: We recommend that future work continue to use and
refine the water-level recorders used in this study. While the device yielded useful data, there were
some signs that it may not hold up under extensive use. Specifically, we noted that the copper-coated
steel welding rods used as guides for the floats and depth indicators were beginning to corrode where
their copper coatings had either been scratched or were plated too thinly. Although it was not a
problem during the short time period the devices were left in place for the pilot study, if the devices are
left in wetland basins for more than 1 year, the corrosion could interfere with the movement of the floats
and depth indicators along the rods and thus cause the devices to give false water-level readings. In
future work, we suggest that commercially available copper-clad welding rods not be used for the guide
rods. Instead, the rods need to be custom made with a thicker copper plating that would resist
scratching and be less likely to have thin spots. Also, we suggest that the PVC casings of the devices
be constructed out of 4-inch (10.2 cm) I.D. pipe instead of the 3-inch (7.6 cm) I.D. pipe used in this
study. This would allow the size of the float to be increased, thus providing greater buoyancy to push
the maximum indicator up the rod and greater weight to push the minimum indicator down if some
minor corrosion of the rod should occur. However, we do not feel that a separate evaluation of the
design modifications would be necessary as was done in this pilot study.
69
-------
Section 6.0
PLANTS AS INDICATORS OF WETLAND CONDITION IN THE PRAIRIE
POTHOLE REGION
Harold A. Kantrud
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
6.1 INTRODUCTION
The purpose of this part of the EMAP pilot study was to determine whether plants
could provide useful site-level indicators of differences between wetlands in lightly-stressed ("good
condition") watersheds and those in heavily-stressed ("poor condition") watersheds. As indicated in
Section 1.0, study sites were selected on the basis of watershed land use. Lightly-stressed watersheds
were considered those dominated by perennial grasses or grass-legume mixtures used for pasture or
hayland or idled under Federal agricultural programs, whereas heavily-stressed watersheds were
considered those dominated by annually-seeded small grain and row crops.
Prairie wetlands are inherently unstable ecosystems because water supplies are variable,
unpredictable, and often largely external. Stresses from agriculture horizontally directed from upper
watersheds toward the wetland centers add to this instability. Agricultural stresses may be direct or
internal, as when the basins themselves are used to raise crops, or indirect or external through siltation
and chemical runoff from the watersheds. Perturbations such as tillage, seeding, fertilizing, and
chemical spraying are common; many wetland watersheds in the Prairie Pothole Region (PPR) have
been used for cropland almost continuously since the late 1800's. In either case, silt and nutrient
loadings increase in affected basins.
The initial direct disturbance to a wetland by cultivation is severe. Cultivation probably affects all
stages in the regeneration cycle of native plants. This cycle is important in maintaining plant species
diversity (Grubb 1977). Tillage equipment severs rhizomes of the native perennial hydrophytes and
overturns and dries the sod; repeated disking and harrowing may follow for a year or more prior to
planting. These operations totally eliminate most of the native plants. After this stage, entire basins of
lesser water permanence are regularly cultivated for crop production or to control weeds whenever
water levels permit. The peripheral zones of wet basins (areas closely related to degree of water
71
-------
permanence and having characteristic assemblages of plants; [Stewart and Kantrud 1971]) are also
regularly cultivated. Thus, these cultivation stresses are, in a sense, predictable and do not allow
recovery of the original native plant community. However, some plants are adapted to repeated
disturbances of bottom substrates and such disturbances may eliminate competitive dominants, thereby
allowing competitive subordinates to occupy the disturbed sites (Wilson and Keddy 1986). In frequently
cultivated wetlands in the PPR, these subordinates consist of a few rapidly-maturing annuals and
relatively short, deep-rooted perennials.
Besides direct tillage, other stressors common to plant communities in frequently cultivated
wetlands are inputs of silt, pesticides, and fertilizers. Silts come from adjacent uplands, but pesticides
and fertilizers can also be directly applied. It is generally unknown whether these stressors have
antagonistic, synergistic, or additive effects (sensu Turner 1985) on the structure and function of these
communities. However, fertilizers normally increase productivity and decrease species richness in most
wetland plant communities studied (Vermeer and Berendse 1983). On the other hand, atrazine-type
herbicides, commonly used on row crops in the region, seem to greatly decrease both production and
species richness, and these decreases may persist at least into the following growing season (James
Richardson, North Dakota State University, pers. comm.; H. A. Kantrud, pers. obs.).
Wetlands of greater water permanence lying in cultivated watersheds are often left idle because
cropping is difficult due to access problems, salinity increases after cultivation, sandy bottoms, or large
boulders or trees are present. These basins are also subject to increased nutrient loadings and many
usually accumulate large amounts of standing dead vegetation. Silt from the adjacent cropped uplands
sometimes is deposited in the peripheral zones to form a barrier or is frequently carried into interior
zones to form a mud delta. Woody plants, especially Salix spp. and Populus spp. also invade idle
wetlands, especially where wet-meadow zones are earlier disturbed by cultivation. Idle coastal marshes
show decreased plant species richness and number of vegetation types present and vegetation
mosaics tend to be coarse-grained (Bakker 1985; Andresen et al. 1990). Idleness also allows formation
of monotypic stands of robust emergents that shade out shorter plants and lowers avian diversity and
abundance (Jones and Lehman 1987; Hellings and Gallagher 1992). For some plants, such shading
can be more important than the effects of herbivory and competition on seedling establishment
(Bergelson 1990). Buildup of litter and organic material from emergent species in prairie wetlands can
reduce water depth or eliminate shallow-water areas (Ward 1942, 1968; Walker 1959; Hammond 1961).
Native plants in the region are adapted to hydrological changes, fire, and herbivory, especially by large
mammals. In pre-agricultural times, these natural forces probably created some sort of normal, but
unknown, homeostatic behavior of the grassland ecosystem. Livestock grazing is currently the dominant
72
-------
land-use practice in grassland-dominated watersheds in the region, although haying is not uncommon.
Pastures in the region nearly always include natural wetland basins that are the most common sources
of livestock water. Nevertheless, many livestock watering facilities have been constructed, some in
natural basin wetlands. Prairie wetlands are basically wet grasslands. Lack of grazing in grasslands is
abnormal; under such situations, plants seemingly dependent on herbivory (obligate grazophiles) can
disappear (McNaughton 1979, 1986). Ratios of standing crop to litter can fall as plant communities age
in the absence of grazing (Bazely and Jeffries 1986). Conversely, livestock grazing, especially in spring
and fall, maintains species richness in meadow grasslands (Smith and Rushton 1994). Grazing in long-
idled salt marshes slowly enhances species diversity and creates fine-grained vegetation mosaics
(Bakker 1985). Grazing by cattle of monodominant stands of Typha glauca in prairie wetlands
decreases live stems, dead stems, and litter (Schultz 1987). Grazing thus may remove much organic
matter and create open water areas where submersed plants flourish. There is a threshold to tolerance
for grazing, however, even in prairie wetlands, because long-term overgrazing and trampling can
reduce the shalbwer zones to nearly bare soil.
Landowners commonly mow and remove emergent vegetation for livestock feed or bedding in
the PPR, especially in watersheds that are seeded to perennial forage crops such as alfalfa. Larger
wetlands in annually-tilled watersheds are also used for hay production. Some native species, such as
Scolochloa festucacea, are considered excellent forage by livestock producers. Others, such as
Phalaris arundinacea, may be seeded in wetland basins. The amount of forage produced and the
wetland zones affected depend on summer or early fall water levels. In basins devoted to forage
production, wet-meadow zones are hayed nearly every year, whereas deep-marsh zones are usually
hayed only after a series of dry years. Most observers agree that long-term use of basins for hayland
tends to increase the abundance of certain emergent hydrophytes (Smeins 1967; Walker and Coupland
1968, 1970; Stewart and Kantrud 1972).
6.2 INDICATORS TESTED
To meet the objectives listed in Section 1.0, I measured
1. abundance and species richness of emergents in temporay, seasonal, and
semipermanent wetlands
73
-------
2. amount of standing dead vegetation and litter in seasonal and semipermanent wetlands
3. abundance of submergents and the ratio of emergent cover to open water in deep-
marsh zones of semipermanent wetlands
4. abundance of metaphytic or planktonic algae in open water areas in deep-marsh zones
of semipermanent wetlands.
Measurements 3 and 4 were dropped from the study because the sample included only four open
water areas in deep-marsh zones; all were found in wetlands in good-condition watersheds, so no
comparisons were possible.
6.3 METHODS
6.3.1 Design
I studied 40 sample wetland basins in 1992 and 36 in 1993-32 of these basins one of the two
years only and 21 in both years (Table 6-1). Field work was conducted in July and the first week of
August because peak standing crops occur during these months in the north-temperate United States
(Bernard 1974). The soil and sediment evaluation researchers accompanied me in the field. I was
provided a large-scale map showing distances between the sample wetland basins. We visited the
basins in a general south-to-north route to help compensate for the approximately two-week difference
in phenology between the southernmost and northernmost study areas. I was provided county road
maps, National Wetland Inventory (NWI) maps, and high-level aerial photographs of the selected 10.4
km2 plots showing the sample wetland basins and the suggested access routes, gates, and parking
locations. I was also provided low-level aerial photographs, taken in mid-June, of each sample wetland
basin (see Section 1 for details). The team carried basin visitation forms and landowner contact forms
giving the landowner's name and telephone number, place of residence, and any special precautions to
be used when visiting the basin.
74
-------
Table 6-1. Sampling design lay-out showing single (X or x) or multiple communities (number)
sampled within deep-marsh (DM), shallow-marsh (SM), or wet-meadow (WM) zones in
basins in good-condition and poor-condition watersheds, 1992-1993.
Good-condition watersheds
Identification
number 1992 1993
Plot Basin DM SM WM DM SM WM
73 29 XXX x x x«
374 100 xxx XXX
374 225 XXX x 2. x
442 301 xxx XXX
156 22 XX XX
363 22 XX X X
363 58 XX xx
442 93 xx XX
442 295 XX xxx
73 86 X x
156 24 x X
156 26 X x
156 42 x X
374 272 2 2_
Good-condition watersheds
Identification
number 1992 1993
Plot Basin DM SM WM DM SM WM
374 65 x X
60 58 XX
60 128 X
249 50 XX
249 86 XX
59 111 X X
396 106 X
396 107 X
396 130 X X
407 67 X
407 109 X
498 146 XXX
498 227 X
498 277 XXX
133 386 X X
407 168 XXX
Total communities 5 13 24 9 13 23
Grand total 87
75
-------
Table 6-1. (continued)
Poor-condition watersheds
Identification
number 1992 1993
Plot Basin DM SM WM DM SM WM
134 140 2. XX
134 165 X XX
134 270 X X
134 406 XX XXX
134 432 X X
442 260 x X
442 261 X X
442 281 xx XX
38 62 X
54 39 X
59 42 X X 2
134 272 X
241 3 X X 2
241 48 2
246 34 X
246 37 X
246 53 XX
133 370 2
133 380 X X
134 158 X
327 72 XX
327 117 X 2
Poor-condition watersheds
Identification
number 1992 1993
Plot Basin DM SM WM DM SM WM
327 147 X X
Total communities 3 8 17 i 8 16
Grand total 53
aData from communities designated by lower case x's and underlined numbers were not used in ANOVAs or for estimating least
squares means for response variables measured at the community level (see methods).
76
-------
6.3.2 Field Methods
6.3.2.1 Watershed and Wetland Classification
Upon arrival at each sample wetland basin, I recorded the time of entry and any landowner
contacts on a basin visitation form. The timing of the surveys generally followed seeding operations and
preceded haying operations.
I visually estimated the current and recent past land uses of the watersheds of each sample
basin. Land-use categories were cultivated, grazed, hayed, burned, and idle. This sometimes required
that I walk to nearby high vantage point(s) around each basin. Watershed land uses were in most
cases similar to those estimated for the recent past (prior growing season). Exceptions were stands of
idle grass that obviously had been grazed or mowed during the previous recent past. There was little
evidence of newly cultivated grassland. For watersheds currently cultivated, I also estimated the
proportional areas of various crops or tillage practices. These catgories included row crop, small grain,
row crop stubble, small grain stubble, weedy fallow, and bare fallow. I also estimated the proportional
areas of the watersheds of each sample basin wetland occupied by annual vegetation (crops and
weeds), perennial vegetation (native grassland or seeded perennials forage crops used for hay), or odd
areas (rockpiles, road right-of-ways, buildings, etc.).
On the low-level aerial photograph of each sample basin wetland, I delineated wetland zones
(Stewart and Kantrud 1971) using a permanent marking pen. A low-prairie zone not recognized as
wetland under the Cowardin et al. (1979) wetland classification occurs around nearly all palustrine and
lacustrine prairie wetlands. This zone is inundated only when water levels are unusually high. Wet-
meadow zones also occur in nearly all palustrine and lacustrine prairie prairie wetlands, the few
exceptions being fens where groundwater seepage is constant, and in small areas along shorelines of a
few semipermanently-flooded or permanently-flooded basins where wave action cuts steep banks along
high-relief shorelines. Wet-meadow zones develop under a temporarily flooded water regime whereby
ponding occurs for a few weeks after spring snowmelt or occasionally for a few days after heavy rains
later in the growing season. Thus for basins lying in cropland, this zone is often available for planting to
spring-seeded crops or for summer or fall tillage for weed control or soil preparation in all but the
wettest years. These conditions often make the outermost edge of the wet-meadow zone difficult to
recognize.
77
-------
Regional shallow-marsh zones are subject to a seasonally flooded water regime whereby
ponding usually occurs for a month or more from spring snowmelt to early summer. During relatively
dry years these zones are also commonly cultivated and planted to spring-seeded crops and tilled in
the fall with the adjacent uplands. Deep-marsh zones ordinarily follow a semipermanently-flooded water
regime whereby surface water normally persists except during a year or more of drought. Fen (alkaline
bog) zones normally saturated by alkaline ground-water seepage sometimes dominate the central areas
of prairie wetlands, but more frequently occur as isolated pockets around the margins of
semipermanently- and permanently-flooded basins.
For each sample wetland basin, I visually estimated the proportional areas of phases (Stewart
and Kantrud 1971) within each zone. Phases reflect changes in water levels and the intensity or
frequency of certain land-use practices. The normal emergent phase is present under normal water
levels and natural unfilled conditions. A natural drawdown phase occurs during periods of low
precipitation. Cultivation of zones results in a cropland tillage phase containing mostly planted crops
and agricultural weeds or a cropland drawdown phase dominated by plants that pioneer on exposed
moist soil after surface water dissipates.
For each sample wetland basin, I visually estimated the proportional area of each zone devoted
to current and recent past (prior growing season) land use practices. Land-use categories were the
same as used for the watersheds.
To fully describe the sample wetland basins I also delineated their other wetlands and within-
basin uplands on the low-level aerial photographs. Other wetlands were dugouts constructed for
watering livestock during drought years. Within-basin uplands were rockpiles or spoils from construction
of the dugouts.
6.3.2.2 Plant Abundance and Species Richness
I delineated plant communities within zones on the low-level photographs using a permanent
marking pen. I numbered each community, and assigned it to a zone (low prairie, wet meadow, shallow
marsh, deep marsh, and fen), phase (natural emergent, natural drawdown, cropland drawdown, and
cropland tillage), and land use (cultivated, grazed, hayed, and idle). Plant communities were considered
vegetation in relatively uniform environments with floristic composition and structure relatively uniform
and distinct from surrounding vegetation (Westhoff and van der Maarel 1973).
78
-------
If wetland basins were subject to different land uses, plant sampling was restricted to that
portion of the basin with predominant land use. Within this area, sampling was further restricted to plant
communities occupying at least 10% of the area (Figure 6-1). I numbered all plant communities, noted
whether they were grazed, hayed, burned, cultivated, or idle, and assigned them to wetland zones and
phases of Stewart and Kantrud (1971). Although vegetation usually forms a virtual continuum around
prairie wetlands, zonation is usually evident (Johnson et al. 1985). In a few instances, drastic water
level increases, cultivation, or rapid crop growth in late June or early July rendered portions of the field
photographs unusable. In those cases, wetland and community boundaries were delineated using
whatever reference points (boulders, haystacks, fencelines) were available.
I used a modified method of Barker and Fulton (1979) for Objective 1 to speed vegetation
sampling. I sampled vegetation along the long axis through the center of each plant community to avoid
edge effects. I paced the long axis, and at each of five roughly equidistant points along the axis I threw
a marker buoy overhead. At its point of impact I centered aim2 collapsible quadrat frame of my own
design (Figure 6-2). I marked quadrat locations on the low-level aerial photographs.
I assigned a Daubenmire (1959) cover class to each macrophyte taxa in the quadrat based on
its shading of the substrate surface. For emergents, substrates were water surfaces if surface water
was present or bottom sediments if surface water was absent. For floating and submerged plants,
substrates were bottom sediments. Midpoint values of the cover classes were used to obtain mean
cover values (n=5) for each plant taxon within the quadrats for each community. Midpoint values were
2.5, 15, 37.5, 62.5, 85, and 97.5. Means of the mean cover values for taxa within the quadrats provided
estimates of the abundance of these taxa among wetland phases. Plant taxa not encountered in the
quadrats were noted while walking between quadrats. Total taxa recorded inside and outside the
quadrants provided a measure of taxa richness for each community. Nearly all plants were identified to
species, but a few were identified only to genus or family or were unidentified. None of the unidentified
plants were identified to species at another sample wetland.
A species list is provided (Appendix 6-1). Mean number of taxa was calculated for each wetland
zone and phase by watershed condition.
79
-------
Overflow (Spill)
ATM From
Another Basin
Community OOZkiNormtiEmrBMtt
community 003 ki Normal bMrgam
PMi»*MlQ«r-fiw*liZM»
iZon*
UIUOTVM Communtt**
Nflttv* Qraaatoid
(Summer Gr«z«d)
Owrtlaw (SpUnAree,
From Another Basin
Low-toy* Photograph
Figure 6-1. Wetland 374-225 (1993) showing sampled and unsampled hydrophyte communities, location of quadrats, and land-use of uplands.
-------
1/4 in. (0.64 cm) O.D. bungee cord (3 meters)
1/2 a. (1.27 can) ID. PVC pipe
,1/2 in. (1.27 cm) I.D. PVC elbows
Figure 6-2. Collapsible quadrat frame.
6.3.2.3 Standing Dead Vegetation and Litter Depth
I visually estimated the percentage of standing dead vegetation in each quadrat. Mean percent
standing dead vegetation (n=5) was calculated for each community.
After I completed the plant sampling, the soils and sediment researchers cored the bottom
substrate at the center of each quadrat with a hand auger and measured litter depth to the nearest cm.
The fresh litter cores were recognizable as undecomposed or partially decomposed fallen vegetation.
The bottom of the litter layer was considered the point where decomposing material changed from fibric
(peat) to hemic (muck) or where plant remains became unrecognizable as such when observed through
a 10X hand lens. The mean depth of the litter layer (n=5 in 1992 and n=3 in 1993) was calculated for
each community. Cores were retained for analyses for the soil study. Another soil sample was bagged
81
-------
in plastic and held on ice for pesticide analysis (described in Section 8.0). I also estimated the
percentages of unshaded bottom and unshaded open water in each quadrat and measured water depth
to the nearest cm at the center of each quadrat. Means for these supplementary variables (n=5) were
also calculated for each community. Means for standing dead vegetation, litter depth, unshaded bottom,
unshaded open water, and water depth were averaged for zones and phases by watershed condition.
My portion of the field work took about 20-25 working days for one person each year. Travel
during field work totalled about 3,200 miles per season.
6.3.3 Analysis
At the end of each field season, I scanned and georeferenced the low-level aerial photos with a
Map and Image Processing System (MIPS) to determine areas of the sampled plant communities. On
each image I classified all polygons as to wetland zone, other wetland (e.g. excavated dugouts for
livestock watering), and included uplands (e.g. spoils or rockpiles), and marked the locations of
quadrats.
We used analysis of variance (ANOVA) techniques to assess the effects of watershed condition
and year on total zone area. Because approximately half of the sample wetland basins were measured
in both 1992 and 1993, the design was one of repeated measures with year serving as the repeated
measures factor. ANOVAs were done separately for each of the five zones; low-prairie, wet-meadow,
shallow-marsh, deep-marsh, and fen.
We also used ANOVA techniques to assess the effects of watershed condition, zone, and year
on all response variables measured at communities within zones (Table 6-2). The sampling design was
a split-plot with repeated measures. Each basin was assumed to be the independent whole-unit, with
zone and community combination being the sub-unit (see Table 6-1). Because some sample wetland
basins were measured in both 1992 and 1993, year served as the repeated measures factor. However,
because of the highly unbalanced design structure (Table 6-1) the three-way interaction effect and least
squares means of wetland condition by zone by year was not fully estimable for the repeated measures
design. Therefore, we randomly deleted one year's data on basins that were used in both 1992 and
1993. This allowed basin to become "nested" within year and wetland condition and thus made the
three-way interaction testable, albeit with slightly less power. We report the least squares means from
this "balancing" approach as all combinations among year, water condition, and zone are estimable.
82
-------
Although not exhaustive, multiple passes were made through the data with a different random selection
each pass. In all passes, ANOVAs yielded similar conclusions. We used Fisher's protected least
significant difference (LSD) to isolate differences in least squares means following significant effects in
the ANOVAs (Milliken and Johnson 1984) where applicable.
All ANOVAs were done using the general linear model procedure (PROC GLM) of SAS (SAS
Inst. 1989). Least squares means (SAS Inst. 1989) were computed and reported when adequate data
was available for ANOVAs. Otherwise, arithmetic means are reported. Effects considered fixed and
random are listed in Table 6-3. For most of the response variables we conducted the ANOVAs both in
the original unit of measurement and using a 1n(Y+1) transformation. We do not report the results of
the transformation, only that we analyzed the response variables in their original scale, (i.e.,
untransformed) and as 1n(Y+1) transformed. The value one was added prior to transformation to
accommodate zero values (Steel and Torrie 1980). Nine zones within a wetland basin had more than
one community (see Table 6-1). We analyzed the data in both scales for two reasons: First, most
biological data tend to follow a log-normal distribution, although we did not test for normality due to
small sample sizes and for reasons described in Johnson and Wichern (1988:155) with respect to
testing statistical normality and second analogous to Conover's (1980:337) recommendation with
respect to using rank transformation and comparing results with untransformed data. However, because
ANOVA results were similar for transformed and untransformed data, we only report the results for
untransformed data; this indicates no gross departures from ANOVA assumptions for untransformed
data. Analyses of the physical and botanical measurements for wetland zones included all wetland
phases. Data were averaged across the five quadrats within each community prior to analysis. With the
exception of the response variable total zone area, the fen and low-prairie areas were considered wet-
meadows. Least squares means in tables are at the highest order interaction for reporting purposes.
6.4 RESULTS
6.4.1 Watershed and Basin Classification
Field inspection revealed that stands of perennial grasses dominated the watersheds of several
sample wetland basins in selected 10.4 km2 plots originally classified as poor condition. Fields in these
watersheds were seeded during the late 1980's or early 1990's under the U.S. Department of
Agriculture's Conservation Reserve Program (CRP). Nevertheless, these sample wetland basins were
83
-------
Table 6-2. Response variables
A. Variables measured at quadrats within communities within zones (all analyses done by first averaging
across quadrats):
1. Plant species and areal cover (Daubenmire) value
2. Water depth (cm)
3. Percent standing dead vegetation
4. Length (cm) of litter core
5. Percent unvegetated (bare) bottom
6. Percent open water
7. Percent vegetation
B. Variables measured at communities within zones:
1. Area of community (ha)
2. Phase of community (normal emergent, cropland drawdown, natural drawdown, cropland tillage,
open water [for deep-marsh zone only])
3. Land use (idle, mowed, cultivated, burned, grazed)
4. Total plant species
5. Total perennial plant species
6. Total annual plant species
7. Total introduced plant species
8. Total native plant species
9. Total perennial introduced plant species
10. Total perennial native plant species
11. Total annual introduced plant species
12. Total annual native plant species
C. Variables measured at zone within wetland basin:
1. Area of zone (ha)
2. Percent of zone in each phase
3. Percent of zone in each land-use (past)
4. Percent of zone in each land-use (current)
D. Variables measured at each wetland basin
1. Percent of wetland basin watershed in annual or perennial cover
2. Percent of wetland basin watershed in each land-use (past)
3. Percent of wetland basin watershed in each land-use (current)
4. Percent of wetland basin watershed in each current crop type
84
-------
Table 6-3. Fixed and random effects in ANOVAs.
Effects
Basin condition
Wetland zone
Year of study
Wetland basin
Community1
Jlype
Fixed
Fixed
Fixed
Random
Random
No. levels
2
3
2
-
-
Levels
Good, poor
Deep-marsh,
(includes fen
1992, 1993
-
-
shallow marsh, wet-meadow
and low prairie)
'Community by zone was considered the sampling unit and randomness assumed.
retained for study because recovery of hydrophyte communities, at least in salt marshes, takes at least
10 to 50 years when soils have been disturbed (Beeftink 1977).
Basin classes and water regimes (palustrine emergent temporarily flooded, palustrine emergent
seasonally flooded, and palustrine emergent semipermanently flooded; Cowardin et al. 1979) shown on
the NWI maps were used for the analyses. Field inspection revealed that several sample wetland
basins may have been misclassified, but they were retained as originally classified for comparisons of
the effects of watershed condition on the response variables. These include a basin (374-272) where
the central or deepest zone was judged to have a saturated water regime (Cowardin et al. 1979) and
three basins (134-432, 156-024, and 396-106) where the central zone may have been low-prairie
(ephemeral wetland of Stewart and Kantrud 1971; non-wetland of Cowardin et al. 1979). These four
zones were considered wet-meadow for analyses of the response variables. All ANOVA tables are in
Appendix 6-2.
ANOVA tests showed no significant differences between wetlands in poor- and good-condition
watersheds in total area (ha) of low-prairie (F151=1.20, p=0.278), wet-meadow zone (F151=1.79,
p=0.187), shallow-marsh zone (F, S1=2.00, p=0.163), deep-marsh zone (F151=1.25 p=0.269), or fen zone
(Fi 51=0.65, p=0.422). Data on other wetlands and included uplands were too sparse to test.
85
-------
6.4.2 Community Characteristics
I sampled vegetation in a total of 144 plant communities during the two-year study.
Communities in the in 76 sample wetland basins visited one or both years included 35 in temporarily-
flooded wetlands, 47 in seasonally-flooded wetlands, and 62 in semipermanently-flooded wetlands
(Tables 6-1, 6-4). Basins in poor-condition watersheds had slightly fewer communities, likely because
cultivation of wet-meadow and shallow-marsh zones during dry years created crop or fallow monotypes.
Grazing, as observed on many wetlands in good-condition watersheds, tends to create more
communities. Semipermanent wetlands in good-condition watersheds had four communities in the
deep-marsh zone in the open-water phase (aquatic bed of Cowardin et al. 1979). Communities with this
combination of zone and phase were not present in semipermanent wetlands in poor-condition
watersheds and so were dropped from the analysis. Data from the remaining 140 communities
(Appendix 6-3) were analyzed.
6.4.2.1 Distribution of Communities Among Wetland Zones and Phases
The analyzed plant communities included 87 (62%) in good-condition watersheds and 53 (38%)
in poor-condition watersheds (Table 6-4). Communities included 80 (57%) in wet-meadow zones, 42
(30%) in shallow-marsh zones, and 18 (13%) in deep-marsh zones. ANOVA results indicated no
significant effects of year, watershed condition, or zone on mean area of communities (ANOVA all
p>0.11). Although not statistically significant, area of communities in good-condition watersheds was, on
average, larger than those in poor-condition watersheds and less variable (Table 6-5).
Total area of the 140 analyzed plant communities was 176.02 ha, including 91% (159.99 ha) in
good-condition watersheds and 9% (16.03 ha) in poor-condition watersheds (Table 6-5). The greater
mean and total area of the communities in good-condition watersheds likely reflects the greater use for
pastures of lands containing larger wetlands.
Of the sampled communities, 111 were in normal emergent phase (150.29 ha), 15 in cropland
tillage phase (2.02 ha), 7 in cropland drawdown phase (1.67 ha), and 7 in natural drawdown phase
(22.04 ha; Table 6-6). About 62% of all sampled communities were in wetlands in good-condition
watersheds. All sampled communities in cropland drawdown and cropland tillage phases were in
wetlands in poor-condition watersheds, whereas all those in the natural drawdown phase were in good-
condition watersheds. Thirteen communities were in the drawdown phase during the relatively dry year
86
-------
Table 6-4. Numbers of sample wetland basins and surveyed wetland plant communities among wetland classes in good-condition and
poor-condition watersheds and mean numbers of communities per basin, EMAP study, Prairie Pothole Region, 1992-1993.
oo
Watershed
condition
Good
Poor
Total
(basin class)8
temporary
seasonal
semipermanent
temporary
seasonal
semipermanent
Basins
8
7
6
7
6
6
40
1992
Communities
13
13
15
7
10
14
72
Basins
6
6
6
6
6
6
36
1993
Communities
7
12
18
8
12
15
72
Mean no. communities
per basin, 1992-1993
1.4
1.9
2.8
1.2
1.8
2.4
'Sample wetland basins classified according to original sample draw.
-------
Table 6-5. Number and least squares means (±SE) of community areas among Stewart and Kantrud
(1971) wetland zones in good-condition and poor-condition watersheds, EMAP study,
Prairie Pothole Region, 1992-1993.
Watershed
condition
Good
Total
Poor
Total
Zone
Wet-meadow
Shallow-marsh
Deep-marsh
Wet-meadow
Shallow-marsh
Deep-marsh
Grand total
No."
24
13
5
42
17
8
3
28
70
1992
Mean area (ha)
0.74 (0
0.59 (0
0.60 (1
0.40 (0
0.28 (1
0.02 (1
.54)
.81)
.88)
.66)
.55)
.63)
No.
23
13
_9
45
16
8
_1
25_
70
1993
Mean area (ha)
2.84 (0
2.20 (0
3.48 (1
0.08 (0
0.42 (0
0.37 (2
.59)
.99)
.08)
.66)
.93)
.65)
"Least squares means are based on fewer samples (see Table 6-1).
of 1992, whereas only a single community was in drawdown during the relatively wet year of 1993.
Because of sparseness of the data (Table 6-6), no attempt was made to assess the effect of condition,
zone, phase, and year on total area.
The estimated mean proportional area of phases for whole sample wetland basins are
presented in Table 6-7 for each zone, year, and watershed condition combination. Because of the
highly skewed nature of the data, no statistical analyses were attempted (i.e., data were mostly either
100% or 0%). The normal emergent phase predominated all zones in basins in good-condition
watersheds. Normal emergent phase varied by watershed condition and year and by watershed
condition and zone. The open water phase was relatively unimportant except in shallow-marsh zones
during the relatively wet year of 1993 where it averaged about 10% of the area of this zone in basins in
poor-condition watersheds.
Proportional areas of zones in the drawdown bare soil and natural drawdown phases did not
vary by watershed condition, year, or zone. These two phases were, of course, most common during
the relatively dry year of 1992. The drawdown bare soil phase tended to be highest in shallow marsh
zones in basins in poor-condition watersheds, whereas the natural drawdown phase tended to be
88
-------
Table 6-6. Number and total area of plant communities sampled among Stewart and Kantrud (1971) phases in good-condition and poor-
condition watersheds, EMAP study, 1992-1993.
Zone
Phase
Wet-meadow
Cropland Drawdown
Cropland Tillage
Natural Drawdown
Normal Emergent
Shallow-marsh
Cropland Drawdown
Natural Drawdown
Normal Emergent
Deep-marsh
Normal Emergent
Total
Number
Good Poor
1992 1993 1992
5
6
4
20 23 6
1
3
10 13 7
593
42 45 28
1993
1
9
-
6
_
_
8
1
25
Total
Good
1992- 1993
_
-
7.69
50.01
_
12.35
37.82
50.12
159.99
area (ha)
Poor
1992- 1993
1.35
2.02
-
4.45
0.32
-
6.29
1.60
16.03
-------
Table 6-7. Means (±SE) of visual estimates of proportional areas of phases of wet-meadow, shallow-marsh, and deep-marsh zones in sample
wetland basins in good-condition and poor-condition watersheds, 1992-1993*.
Proportion of zone areas
Zone
Phase
Wet-meadow
Normal emergent
Open water
Drawdown bare soil
Natural drawdown
Cropland drawdown
Cropland tillage
Shallow-marsh
Normal emergent
Open water
Drawdown bare soil
Natural drawdown
Cropland drawdown
Cropland tillage
Deep-marsh
Normal emergent
Open water
Drawdown bare soil
Natural drawdown
Cropland drawdown
Cropland tillage
Good-condition
1992
81.6
0.0
4.6
13.4
0.0
0.1
79.2
0.0
7.1
13.7
0.0
0.0
91.9
5.6
0.0
6.7
0.0
0.2
(6.1)
(0.0)
(2.7)
(4.1)
(0.0
(4.3)
(7.6)
(0.0)
(3.4)
(4.7)
(0.0)
(0.0)
(11.9)
(3.9)
(0.0)
(6.5)
(0.0)
(8.5)
1993
96.3
0.0
0.0
4.3
0.0
1.8
99.9
0.1
0.0
4.3
0.0
0.0
98.5
3.3
0.0
1.5
0.0
0.2
(6.2)
(0.0)
(0.0)
(4.1)
(0.0)
(4.3)
(7.8)
(2.5)
(0.0)
(4.1)
(0.0)
(0.0)
(9.2)
(3.0)
(0.0)
(5.3)
(0.0)
(7.6)
Poor-condition
1992
27.2 (7.2)
1.8 (2.3)
0.0 (0.0)
0.0 (0.0)
26.1 (3.7)
44.9 (5.0)
83.5 (10.5)
0.0 (0.0)
7.2 (4.8)
0.0 (0.0)
3.5 (5.5)
3.5 (7.4)
83.6 (16.0)
0.1 (5.2)
1.0 (7.3)
0.0 (0.0)
3.3 (8.4)
5.9 (11.7)
1993
13.7
3.4
0.0
0.0
0.0
84.0
83.6
10.0
0.0
0.0
2.0
2.9
82.5
1.8
0.7
0.0
3.7
5.5
(7.6)
(2.4)
(0.0)
(0.0)
(0.0)
(5.3)
(9.9)
(3.2)
(0.0)
(0.0)
(5.1)
(7.0)
(26.2)
(8.7)
(12.4)
(0.0)
(14.2)
(19.8)
"Based on visual estimates and unweighted for area of individual zones or phases.
-------
highest in wet-meadow and shallow-marsh zones in basins in good-condition watersheds. Proportions
of cropland drawdown and cropland tillage phase in the sample basins varied by watershed condition,
year, and zone. Greatest proportions of cropland drawdown zone occurred in wet-meadow zones in
basins in poor-condition watersheds during the dry year of 1992, whereas greatest proportions of
cropland tillage phase were found in wet-meadow zones of basins in poor condition watersheds during
the wetter year of 1993.
6.4.2.2 Land Use of Wetland Zones
Wet-meadow, shallow-marsh, and deep-marsh zones in the sample wetland basins were
subject to three major land uses as well as idle conditions that strongly reflected their water regimes
and watershed conditions (Table 6-8). No burned wetlands or zones of wetlands were included in the
sample, but this land use is practiced in some areas of the region.
Land use of wetland basins in the recent past may better reflect long-term use than observed
current use because wetlands are often hayed and grazed in late summer or early fall. Higher
proportions of zones of sample wetland basins in poor-condition watersheds showed past cultivation, as
evidenced by furrows and unearthed rocks and boulders left by tillage equipment. Wet-meadow zones
were cultivated to a greater extent in the past than zones of greater water permanence. Past grazing of
basins, as evidenced by trails and old cattle dung, in good-condition watersheds was greater than that
of basins in poor-condition watersheds. Proportions of zones estimated to be mowed or idle in the past
did not vary by watershed condition.
Higher proportions of zones of sample wetland basins in poor-condition watersheds were
currently cultivated. Wet-meadow zones were cultivated to a greater extent than zones of greater water
permanance. Basins in good-condition watersheds were currently grazed to a greater extent than
basins in poor-condition watersheds. Proportions of zones currently mowed or idle did not vary by
watershed condition.
6.4.2.3 Land Use of Watersheds
Mean proportional areas of major watershed cover types were obtained from visual estimates of
the catchment areas of the sample wetland basins (Table 6-9). These estimates reflected the technique
used to select the sample wetland basins. Poor-condition watersheds contained higher amounts of
annual crop plants (F, 51=126.83, p=0.001), whereas perennial vegetation, mostly native and seeded
91
-------
Table 6-8. Means (±SE) of proportion of zones of basins in good-condition and poor-condition watersheds subjected to various current and
recent past land uses, EMAP study, Prairie Pothole Region, 1992-1993*.
to
ro
Proportion of zones
Good-condition watersheds
Zone
Land use
Wet-meadow
Cultivated
Grazed
Hayed
Idle
Shallow-marsh
Cultivated
Grazed
Hayed
Idle
Deep-marsh
Cultivated
Grazed
Hayed
Idle
Current
1992
Tr.
59
4
36
0
54
Tr.
44
0
41
Tr.
56
b(4)
(8)
(2)
(9)
(0)
(10)
(2)
(11)
(0)
(14)
(4)
(16)
1993
2 (4)
61 (8)
Tr.(2)
36 (9)
0 (0)
55(10)
Tr.(2)
44(11)
0 (0)
47(11)
Tr.(3)
51(12)
Past
1992
4 (6)
73 (8)
8 (6)
15 (9)
3 (6)
72(10)
8 (8)
19(10)
3 (9)
75(16)
3(11)
22(16)
Poor-condition watersheds
Current
1993
5 (6)
63 (8)
14 (6)
20 (9)
2 (7)
55(10)
20 (8)
24(11)
3 (7)
46(12)
6 (9)
47(12)
1992
63 (5)
0 (0)
0 (0)
37(10)
0 (0)
0 (0)
0 (0)
84(14)
0 (0)
0 (0)
0 (0)
78(21)
1993
87 (5)
0 (0)
0 (0)
12(11)
0 (0)
0 (0)
0 (9)
89(14)
0 (0)
0 (0)
0 (0)
81(33)
Past
1992
71 (7)
0 (0)
8 (7)
21(10)
52 (8)
0 (0)
12(10)
44(14)
41(12)
0 (0)
7(15)
69(21)
1993
80 (7)
0 (0)
9 (8)
9(11)
18 (8)
0 (0)
28(10)
54(13)
28(17)
0 (0)
12(24)
67(34)
'Based on visual estimates and unweighted for area of individual zones.
"Tr.=<0.5%
-------
grasses and forage crops, was far more important in good-condition watersheds
(F1S1=162.44, p=0.001). Other cover types did not vary by watershed condition (F151 1.42, p=0.24).
Mean current and recent past land use practices of the watersheds of the sample wetland
basins, based on visual estimates, also strongly reflected the technique used to select the sample
wetland basins (Table 6-10). Amounts of past and currently-cultivated land tended to be higher in poor-
condition watersheds and amounts of past and currently-grazed land were higher in good-condition
watersheds. These conditions likely had been stable for many years, as little land clearing, except for
some minor removal and tillage of fencelines, was evident during the study. Amount of idle land tended
to be higher in good-condition watersheds in the recent past and currently. Most land in a single,
entirely idle watershed was enrolled in the CRP. The proportions of watersheds devoted to hayland did
not vary by watershed condition. Some areas hayed in June during the dry year of 1992 were idle
during the July surveys the following wet year. Current cropping patterns for agricultural land in the
watersheds of the sample wetland basins are shown in Table 6-11. Amounts of seeded small grains
increased and amounts of bare fallow decreased dramatically during the wetter year of 1993, especially
in poor-condition watersheds. Proportions of watersheds planted to row crops and small grains and in
bare tilled fallow were greater in poor-condition watersheds than in good-condition watersheds.
Proportions of small grain stubble and weedy fallow did not vary by watershed condition. No row crop
stubble occurred on the watersheds of the sample wetland basins.
Table 6-9. Means (± S.E.) proportion of major watershed cover types in good-condition and poor-
condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993'.
Proportion
of plant cover types in basin watersheds
Good-condition
Cover type
Annuals
Perennials
Other"
1992
12.5
82.4
0.2
(1
(4
(0
.6)
.0)
.4)
1993
15
85
0
.5 (1.6)
.4 (4.2)
.9 (0.4)
1992
93.7
1.7
4.6
Poor-condition
1993
(1
(4
(0
.9)
.9)
.4)
93.7
1.7
4.6
(2
(5
(0
.3)
.8)
.5)
'Based on visual estimates and unweighted for area of individual cover types.
Includes roads, farmsteads, other built-up areas, gravel pits, and rockpiles.
93
-------
Table 6-10. Means (± S.E.) proportion of watersheds in current and recent past land use practices for sample wetland basins in good-condition
and poor-condition watersheds EMAP study, Prairie Pothole Region, 1992-1993*.
Proportion of watersheds
Good-condition watersheds
Current
Land use
Cultivated
Grazed
Hayed
Idle
1992
11
48
11
30
(4)
(6)
(3)
(5)
1993
11
58
3
28
(4)
(6)
(3)
(5)
Past
1992
15
51
10
24
(4)
(6)
(3)
(5)
1993
14
59
14
12
(4)
(6)
(3)
(5)
Poor-condition watersheds
Current
1992
98
Trb
Tr
1
(5)
(7)
(4)
(6)
1993
98
Tr
Tr
1
(5)
(8)
(4)
(7)
Past
1992
98
Tr
Tr
1
(4)
(7)
(4)
(6)
1993
98
Tr
Tr
1
(5)
(8)
(4)
(7)
"Based on visual estimates and unweighted for area of individual watersheds.
Tr.=<0.5%.
-------
Table 6-11. Means (± S.E.) proportion of currently raised crops on annually tilled land in watersheds in
good-condition and poor-condition watersheds, EMAP study, Prairie Pothole Region 1992-
1993V '
Proportion of basin
Current cropland
land use
Row crops
Small grains
Small grain stubble
Weedy fallow
Bare fallow
watersheds
Good-condition
1992
9.8
6.8
0.0
0.0
0.2
(4.0)
(5.1)
(0)
(0)
(4.8)
1993
6
8
0
0
0
.8
.7
.0
.5
.0
(3.9)
(5.0)
(0)
(0.3)
(0)
Poor-condition
1992
40.
18.
9.
0.
31.
8
1
8
0
1
(4.7)
(6.0)
(3.9)
(0)
(5.7)
1993
30.
78.
0.
0.
0.
2
7
0
0
0
(5.5)
(7.1)
(0)
(0)
(0)
°Based on visual estimates and unweighted for area of individual cropfields or crop types.
6.4.2.4 Physical Measurements in Communities
Physical and botanical measurements derived from quadrats were summarized for the three
wetland zones (Tables 6-12 to 6-16) and tested with ANOVA.
Water depth varied with year and zone (F236=4.44, p=0.019; Table 6-12), but there were no
significant watershed condition effects. Increased precipitation resulted in higher water depths in 1993
(F149=6.49, p=0.014). Depths were higher in zones of greater water permanence (F236=20.54, p=0.001).
Percent of standing dead vegetation did not vary by year and watershed condition (F, 49=0.35,
p=0.555; Table 6-13). However, greater amounts of standing dead vegetation were found in zones of
greater water permanence (F236=6.78, p=0.001).
Depth (cm) of litter varied with watershed condition, zone, and year (F236=4.70, p=0.015; Table
6-14). Depth of litter was higher in zones of greater water permanence and in zones of sampled
wetlands in poor-condition watersheds during the dryer year of 1992. Effects of watershed condition
alone were non-significant.
Percent unvegetated bottom varied with year (F149=4.53, p=0.038) and condition (F149=10.03,
p=0.003; Table 6-15). Greater amounts of unvegetated bottom were found in sample wetland basins in
poor-condition watersheds and lesser amounts occurred in all wetlands during the wetter year of 1993.
95
-------
Table 6-12. Least squares means (±SE) water depth (cm) in plant communities' in wet-meadow,
shallow-marsh, and deep-marsh zones of sample wetland basins in good-condition and
poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993.
Watershed
condition
Water depth (cm)
Year
Zones
1992
1993
Good
Poor
Wet-meadow
Shallow-marsh
Deep-marsh
Wet-meadow
Shallow-marsh
Deep-marsh
2
7
26
0
2
28
.2
.9
.3
.4
.9
.7
(2
(4
(9
(3
(8
(8
.8)
.2)
.8)
.4)
.0)
.4}
3
23
44
8
37
38
.6
.4
.4
.1
.0
.8
(3
(5
(5
(3
(4
(13
.0)
.1)
.6)
.4)
.8)
.6)
"Sample sizes for communities as in Table 6-5; least squares means based on (ewer (Table 6-1).
Table 6-13. Least squares means (±SE) of percent standing dead vegetation in plant communities* in
wet-meadow, shallow-marsh, and deep-marsh zones of sample wetland basins in good-
condition and poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993.
Watershed
condition
% standing dead vegetation
Year
Zones
1992
1993
Good
Poor
Wet-meadow
Shallow-marsh
Deep-marsh
Wet-meadow
Shallow-marsh
Deep-marsh
6.6
8.0
17.0
2.1
0.1
8.0
(1.1)
(1.6)
(3.7)
(1.3)
(3.1)
(3.2)
2.0
3.1
10.5
1.5
0.6
3.8
(1.2)
(2.0)
(2.1)
(1.3)
(1.9)
(5.2)
"Sample sizes for communities as in Table 6-5; least squares means based on fewer (Table 6-1).
96
-------
Table 6-14. Least squares means (±SE) of litter depth (cm) in plant communities* in wet-meadow,
shallow-marsh, and deep-marsh zones in sample wetland basins in good-condition and
poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993.
Watershed
condition Zones
Good
Wet-meadow
Shallow-marsh
Deep-marsh
Poor
Wet-meadow
Shallow-marsh
Deep-marsh
Litter depth (cm)
Year
1992
0
0
0
1
2
6
.3
.2
.0
.0
.7
.7
(0
(0
(1
(0
(1
(1
.3)
.5)
.2)
.4)
.0)
.0)
0
1
1
0
0
0
1993
.2
.7
.3
.0
.0
.0
(0
(0
(0
(0
(0
(1
.4)
.6)
.7)
.4)
.6)
.7)
'Sample sizes for communities as in Table 6-5; least squares means based on fewer (Table 6-1),
Table 6-15. Least squares means (±SE) of percent unvegetated bottom in plant communities* in wet-
meadow, shallow-marsh, and deep-marsh zones of sample wetland basins in good-
condition and poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993.
Watershed
condition
% unvegetated bottom
Year
Zones
1992
1993
Good
Wet-meadow
Shallow-marsh
Deep-marsh
2.0 {2.4}
10.0 (3.5)
12.4 (8.3)
0.9 (2.6)
1.0 (4.3)
0.5 (4.7)
Poor
Wet-meadow
Shallow-marsh
Deep-marsh
46.5
b
(2.8)
15.2 (2.8)
19.4 (4.0)
17.3 (11.6)
'Sample sizes for communities as in Table 6-5; least squares means based on fewer (Table 6-1).
"Least squares means were poorly estimated for poor condition in 1992; observed means are 43.7, 0.0, and 0.0 for wet-meadow,
shallow-marsh, and deep-marsh, respectively.
97
-------
Table 6-16. Least squares means (±SE) of percent open water in plant communities* in wet-meadow,
shallow-marsh, and deep-marsh zones of sample wetland basins in good-condition and
poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993.
Watershed
condition
% open water
Year
Zones
1992
1993
Good
Poor
Wet-meadow
Shallow-marsh
Deep-marsh
Wet-meadow
Shallow-marsh
Deep-marsh
4
8
1
0
2
9
.3
.3
.7
.6
.8
.1
(3
(5
(12
(4
(10
(10
.6)
.4)
.5)
.3)
.2)
.7)
3
35
34
18
36
40
.2
.6
.6
.4
.5
.4
(3
(6
(7
(4
(6
(17
.9)
.5)
.1)
.3)
.1)
.5)
"Sample sizes for communities as in Table 6-5; least squares means based on fewer (Table 6-1).
Greater amounts of open water occurred during the wetter year of 1993 (F,|49=5.12, p=0.028;
Table 6-16), and in zones of greater water permanence (F236 = 6.06, p=0.0328), but effects of
watershed condition alone were non-significant.6-16), and in zones of greater water permanence
(F236 = 6.06, p=0.0328), but effects of watershed condition alone were non-significant.
Physical and botanical measurements derived from quadrats were also summarized for wetland
phases within zones of the sample wetland basins (Table 6-17), but data were too sparse to test with
ANOVA. In the dry year of 1992, communities in drawdown phases were common. Those in the natural
drawdown phase found in wet-meadow and shallow-marsh zones of basins in good-condition
watersheds had more standing dead vegetation and less unvegetated bottom than the ones in these
same zones in the cropland drawdown phase of basins poor-condition watersheds.
6.4.2.5 Botanical Measurements of Communities
A total of 298 major (within quadrats) and minor (observed outside quadrats) plant "taxa" was
recorded (Appendix 6-1), including 217 wetland pteridophytes and spermatophytes (73%) listed for the
north plains (Reed 1988), 50 upland spermatophytes (17%) listed in the National List of Scientific Plant
Names (USDA 1982), and 31 (10%) other "taxa." These were certain non-vascular plants including the
macroalgae Chara spp., two liverworts (Riccia fluitans and Ricciocarpus natans), the aquatic moss
98
-------
Drepanocladus spp., and unidentified plants (e.g. Gramineae unidentified) seen only in early growth
stages. All vascular plants were classified as to life history (annual or biennial, perennial, native,
introduced, or adventive) whenever possible.
Total taxa recorded was higher in all zones of sample wetland basins in good-condition
watersheds throughout the study (Table 6-18). Ratios of total taxa recorded in good-condition versus
poor-condition watersheds varied from a low of about 1.6:1 in wet-meadow zones to a high of about
3.4:1 in deep-marsh zones. The greatest numbers of taxa were recorded in communities in wet-
meadow zones in good-condition watersheds (173) in 1992 and lowest (8) in the single community in a
deep-marsh zone in a poor-condition watershed studied in 1993. Highest mean taxa richness during
the study was recorded in wet-meadow zones of basins in good-condition watersheds during 1992.
Lowest mean taxa richness was found that same year in deep-marsh zones of wetlands in poor-
condition watersheds. When unadjusted for community size, taxa richness varied by zone (F2 38=17.35,
p=0.0001) and watershed condition (F249=3.94, p=0.053), with richness higher in wet-meadow zones
and shallow-marsh zones in good condition than in similar zones of poor condition. As a partial test of
effects of community size, the 17 communities in good-condition watersheds larger than the largest
communities in poor-condition watersheds (1.43 ha) were eliminated from the data set. Effects of zone
remained significant (F228=15.09, p=0.0001), whereas effects of watershed condition was marginally
significant (F146=2.59, p=0.114). The only significant correlation between taxa richness and community
size was for communities of deep-marsh zone within good-condition watersheds (r=0.77; p=0.0013).
Although data were too sparse to conduct statistical tests, total taxa recorded was also
uniformly greater in comparable phases of zones of sample wetland basins in watersheds in good
condition versus those in poor condition (Table 6-19). For normal emergent phases, ratios of total taxa
recorded in good-condition versus poor-condition watersheds varied from a low of about 1.7:1 in
shallow-marsh zones to a high of about 3.4:1 in deep-marsh zones. Greatest numbers of taxa were
recorded in the normal emergent phase of wet-meadow zones in good-condition watersheds in 1993
(166) and lowest in the normal emergent phase of the single deep-marsh zone in a poor-condition
watershed studied in 1993 (8).
The highest mean taxa richness (15.25 taxa/community) occurred in communities in the natural
drawdown phase of wet-meadow zones in sample wetland basins in good-condition watersheds. Mean
taxa richness was higher in the normal emergent phase of wet-meadow zones of basins located in
poor-condition watersheds than in good-condition watersheds during both years. Lowest mean taxa
richness (3.2-4.1 taxa/community) was recorded in the cropland tillage phase of wet-meadow zones in
99
-------
basins in poor-condition watersheds. The greatest number of total (major and minor) taxa (49) was
found in the normal emergent phase of the wet-meadow zone of a grazed semipermanent wetland in a
good-condition watershed, whereas the fewest (0) were in the cropland tillage phase of a wet-meadow
zone of a temporary wetland in a poor-condition watershed. The information on plant taxa could be
biased because we took an equal number of samples in each community, regardless of its area, during
this rapid field evaluation. We know that larger communities have more species, so our crude test
(reanalyzing after eliminating large communities) should have some meaning. Many more analyses,
including the construction of diversity-area curves, could have been done.
Perennial native plants dominated all zones of sample wetland basins in both good-condition
and poor-condition watersheds (Table 6-20). Greatest number of these plants (116 taxa) were found in
communities in wet-meadow zones of sample wetland basins in good-condition watersheds during the
wetter year of 1993, whereas fewest (7 taxa) were found that same year in communities in deep-marsh
zones of basins in poor-condition watersheds. ANOVA tests showed that mean number of native
perennials varied by zone and watershed condition (F236=2.79, p=0.075). Wet-meadow zones in
good-condition watersheds had greater numbers of native perennials than those in poor-condition
watersheds. This relation also held when the 17 communities in good-condition watersheds larger than
the largest communities in poor-condition watersheds were eliminated from the data set (F228=2.76,
p=0.081). Ratios of native perennials to introduced perennials varied by watershed condition and zone
when adjusted as above for community size (F228=3.37, p=0.049). With this adjustment, the ratio of
native perennials to introduced perennials was marginally greater in good-condition watersheds than in
poor condition watershed for wet-meadow zones only. Ratios of native annuals to introduced annuals
did not vary by zone or watershed condition.
More annuals, both native and introduced, were generally found during the drier year of 1992
(Table 6-20). This likely reflects the increased occurrence of drawdown species that pioneer on bare
mud flats and upland species that invade wetlands during drought. Greater numbers of introduced
perennials were found in basins in good-condition watersheds.
The effects of various land use practices were evident in the life history and origin of the
species in wet-meadow zones (Table 6-21). Those in the normal emergent phase in both good-
condition and poor-condition watersheds were dominated exclusively by native and a few introduced
perennials, especially Poa pratensis and Agropyron repens. However, the other dominants differed
greatly. Those in emergent wet meadows in good-condition watersheds were mostly fine-stemmed
grasses and sedges and a few forbs indicative of long-term grazing or mowing, whereas those in
100
-------
Table 6-17. Mean (n=5 quadrats/community) physical and vegetatlonal features of emergent plant communities* in phases of zones of sample
wetland basins in wetlands in good-condition and poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993.
Condition
Zone
Phase
Good
Wet-meadow
Normal emergent
Natural drawdown
Shallow-marsh
Normal emergent
Natural drawdown
Deep-marsh
Normal emergent
- Poor
_*
Wet-meadow
Normal emergent
Cropland drawdown
Cropland tillage
Shallow-marsh
Normal emergent
Cropland drawdown
Deep-marsh
Normal emergent
Water
depth (cm)
1992 1993
1.9 3.5
0.0
13.1 29.4
0.0
35.6 44.7
0.1 10.8
0.4 0.0
0.7 8.6
1.6 31.9
0.0
29.6 16.3
% standing
dead veg.
1992
6.8
4.1
3.1
9.5
16.2
6.2
tr.b
0.0
3.9
0.0
13.8
1993
1.7
-
2.0
-
9.9
2.5
0.0
0.0
1.1
-
10.0
Litter
depth(cm)
1992 1993
0.4 0.3
0.0
0.1 1.0
0.3
1.2 1.5
0.5 0.0
0.0 0.0
0.0 0.0
3.1 0.0
3.1
10.5 0.0
% unvegetated
bottom
1992
1.0
3.3
5.2
23.4
2.4
0.0
80.9
53.6
0.9
52.7
0.0
1993
0.6
-
1.0
-
0.0
0.0
69.1
30.6
3.9
-
0.0
% open
water
1992 1993
3.7 3
0.0
13.5 31
0.0
9.9 34
0.0 14
0.0 0
5.3 22
0.2 26
0.0
6.7 13
.4
-
.2
-
.8
.0
.0
.5
.3
-
.0
"Sample sizes for communities sampled same as in Table 6-5.
"Tr.=<0.05%
-------
Table 6-18. Total plant taxa and least squares means (±SE) taxa richness for communities in wet-
meadow, shallow-marsh, and deep-marsh zones in sample wetland basins in good-
condition and poor-condition watersheds, EMAP study, Prairie Pothole Region, 1992-1993*.
1992
Good-condition
Zone
Total
Wet-meadow
Shallow-marsh
Deep-marsh
communities
nb
24
13
5
42
No.
taxa
173
90
31
Mean taxa
richness
23.8 (1.7)
12.0 (2.5)
2.4 (5.9)
1993
n
17
8
3
28
Good-condition
Zone
Total
Wet-meadow
Shallow-marsh
Deep-marsh
communities
nb
23
13
9
45
No.
taxa
166
74
47
Mean taxa
richness
25.4 (1.8)
12.8 (3.1)
7.4 (3.4)
n
16
8
1
25
Poor-condition
No.
taxa
104
47
15
Mean taxa
richness
11.1 (2.0)
1.7 (4.8)
0.0 (5.1)
Poor-condition
No.
taxa
89
49
8
Mean taxa
richness
14.8 (2.0)
8.2 (2.9)
9.5 (8.3)
'Means unadjusted for community area. Some taxa common to more than one zone.
bn=Number of communities; least squares means based on fewer (see Table 6-1).
similar habitats in poor-condition watersheds were mostly coarse grasses and woody plants indicative
of past disturbance by tillage or possibly siltation.
Although the natural drawdown phase found in wet meadows in watersheds in good-condition
and the cropland drawdown phase found under similar water regimes in poor-condition watersheds
were dominated by mixtures of native and introduced perennials and native and introduced annuals,
there was a preponderance of small native annuals that germinate on exposed bare soil in the poor-
condition watersheds. Most important plants of the cropland tillage phase were introduced annuals,
including at least four species of annual small grains and row crops.
6.5 EVALUATION
Cultivation of various emergent wet-meadow and shallow-marsh communities during dry years
seems to create coarser grained vegetation mosaics with fewer communities as old annular stands of
102
-------
Table 6-19. Number of plant taxa recorded in communities among phases in sample wetland basins in good-condition and poor-condition
watersheds, EMAP study, Prairie Pothole Regions, 1992-1993*.
Zone
Phase
Wet-meadow
Cropland Tillage
Cropland Drawdown
Natural Drawdown
Normal Emergent
Shallow-marsh
Cropland Drawdown
Natural Drawdown
Normal Emergent
Deep-marsh
Normal Emergent
Total communities
No. of taxa
1992
Good
nb
_
_
4
20
-
3
10
_5
42
No.
taxa
_
-
61
150
-
41
76
31
Poor
n
6
5
-
6
1
-
7
3
28
No.
taxa
19
33
_
74
16
-
41
15
No. of taxa
1993
Good Poor
No. No.
n taxa n taxa
9 37
1 14
-
23 166 6 68
-
_
13 74 8 49
_9, 47 _i 8
45 25
"Some taxa common to more than one zone or phase.
bn=Number of communities
-------
Table 6-20. Total numbers of perennial, annual (includes biennial), native, and introduced plant taxa in
communities in wet-meadow, shallow-marsh, and deep-marsh zones of sample wetland
basins in good-condition and poor-condition watersheds, EMAP study, Prairie Pothole
Region, 1992-1993.
1992
Total number of plant taxa
1993
Watershed condition
Watershed condition
Zone
Life History Status
Good
Poor
Good
Poor
Wet-meadow
Perennial-Native
Perennial-Introduced
Annual-Native
Annual-Introduced
Life history unknown
Shallow marsh
Perennial-Native
Perennial-Introduced
Annual-Native
Annual-Introduced
Life history unknown
Deep-marsh
Perennial-Native
Perennial-Introduced
Annual-Native
Annual-Introduced
Life history unknown
109
17
22
12
13
45
9
24
6
6
16
2
8
0
5
59
10
19
e
8
28
5
9
2
3
11
2
0
0
2
116
16
14
8
9
48
7
11
3
5
29
5
3
1
9
43
11
19
9
7
29
5
10
3
1
7
0
1
0
0
weedy annuals, drawdown species, or early successional vegetation are converted to crops or fallowed.
Yet hydrophyte communities in good-condition watersheds tended to be larger and probably reflect the
greater use for pastures of lands containing larger wetlands. Some larger wetlands in poor-condition
watersheds were also used for hay or pasture. If wet-meadow zones are left uncultivated and if siltation
is not severe, these wetlands appear similar to those in good-condition watersheds. However, nearly all
basins in poor-condition watersheds are cultivated for crop production or weed control whenever
bottoms are dry.
Greater percentages of standing dead vegetation in the deeper, more permanent zones likely
reflect a reduced accessibility of these zones to livestock and farm equipment. Sample sizes were too
small (n=3 and n=1 for 1992 and 1993, respectively) to detect the greater amounts of dead vegetation
and litter expected in deep-marsh zones in poor-condition watersheds because of siltation and lack of
grazing, cultivation, or other mechanisms that reduce plant biomass. Agricultural and pastoral
104
-------
Table 6-21. Mean areal cover values and life history status in the prairie pothole region for the 10 most abundant species in communities in
phases of wet-meadow zones of sample basin wetlands in good-condition and poor-condition watersheds, EMAP study, 1992-1993.
Seeded crop plants are marked with an asterisk^).
Taxa
Carex lanuginosa
Calamagrostis inexpansa
Hordeum jubatum
Glyceria striata
Poa palustris
Potentilla anserina
Carex praegracilis
Agropyron caninum
Symphorbarpos occidentalis
Solidago canadensis
Carex aquatilis
Polygonum amphiblum
Puccinellia nuttalliana
Ambrosia psilostachya
Spartina pectinata
Phalaris arundinacea
Salix exigua
Cornus stolonifera
Salix amygdaloides
Calamagrostis canadensis
Boltonia asterioides
Stachys palustris
Eleocharis acicularis
Limosella aquatica
Poa pratensis
Medicago sativa
Bromus inermis
Life
history*
NP
NP
NP
NP
NP
NP
NP
NP
NP
N
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
NP
IP
IP
IP
Good
Normal
emergent
1992 1993
5.6 7.7
5.6 5.0
4.3 5.2
4.2 4.2
6.7
4.4
4.2
4.2
4.0
3.9
3.7
2.9
_
-
-
-
-
_
-
-
-
-
-
-
8.6 14.7
4.4
3.6
% areal cover
Watershed condition
Poor
Natural Normal Cropland Cropland
drawdown emergent drawdown tillage
1992 1992 1993 1992 1993 1992 1993
______
______
30.6 ______
______
6.6 - -
- ______
______
______
______
4.4 - - -
______
2.9 8.1 6.8 1.5 - 0.1 3.2
3.9 ______
3.6 -____-
10.5 9.8 - - -
5.1 16.3 - - - -
9.2 - - -
5.0 - - - - -
4.6 _____
9.2 - - - -
2.2 - -
1.8
2.0 - -
0.6 - -
8.8 5.2 - - -
______
______
-------
Table 6-21 (continued)
Taxa
Agropyron repens
Melilotus spp.
Amaranthus retroflexus
Conzya canadensis
Eleocharis engelmannii
Gratiola neglects
Veronica peregrina
Polygonum lapathffolium
Plagiobothrys scouleri
Potentilla norvegica
Senecio congestus
Artemisia biennis
Lactuca serriola
Kochia scoparia
Bromus japonicus
Triticum aestivum*
Chenopodium album
Zea mays*
Glycine max*
Setaria spp.
EchinochJoa crusgalli
Avena saliva
Sinapsis arvenis
Hordeum vulgare
Carex, unidentified
Forb, unidentified No. 1
Forb, unidentified No. 4
Life
history8
IP
IP
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
IA
IA
IA
IA
IA
IA
IA
!A
IA
IA
IA
IA
-
-
™
% areal cover
Watershed condition
Good Poor
Normal Natural Normal Cropland
emergent drawdown emergent drawdown
1992 1993 1992 1992 1993 1992 1993
20.0 16.2
- _ _
5.1 - 1.3
3.0 -
0.8 12.5
- - - - 1.0 2.5
5.0
- - - 3.5
2.5
- - - - - 1.6
- - - - 0.4
- - _
15.8 - -
- 9.5 -
6.8 -
- - - _
- - _
- - _
- - -
- 2.5
- - - - 10.2
- - - 7.0
- 3.0
_
3.1 - -
- - 0.4 3.5
- 1.0
Cropland
tillage
1992 1993
0.4
0.1
0.3
_
0.6
0.2
-
-
-
-
-
0.8
_ _
0.2
-
19.2 37.8
0.2 0.2
13.2
7.9
6.9
3.2
3.3
2.9
7.7
-
-
— —
"Codes: NP=native perennial; NA=native annual; IP=introduced or adventive perennial; lA-introduced or adventive annual or biennial.
-------
operations tended to reduce standing dead vegetation in the less permanent zones of both watershed
types. Livestock grazing pressure in these zones in basins in good-condition watersheds probably was
insufficient to greatly reduce standing dead vegetation. I also noted that sample sizes of deep-marsh
zones in poor-condition watersheds were very small.
As with standing dead vegetation, litter core lengths were naturally higher in zones of greater
water permanence, probably because they were less accessible to machinery and livestock. Greater
biomass production could also be another factor because the more permanent zones usually support
taller, more robust plant species. Nevertheless, the presence of surface water limits access by
machinery more than cattle. Thus litter depth among zones varied significantly only in basins in poor-
condition watersheds because these basins were usually ungrazed. There were no significant effects
due to watershed condition alone. The irregular destruction of litter by machinery in basins in poor-
condition watersheds likely was not much greater than that caused by the often season-long livestock
hoof action and herbivory that compress or reduce the litter layer in grazed basins. A single pass by
tillage equipment often tears narrow openings in vegetation, and can leave much of the litter layer
intact. Also, basins in poor-condition watersheds often receive inputs of fertilizer from their adjacent
cropped uplands that could increase plant biomass in areas where root systems are not directly
destroyed by tillage.
The greater percentages of unvegetated bottom found in wet-meadow zones of wetlands in
poor-condition watersheds undoubtedly reflect the effects of cultivation. Communities in these
watersheds tended to have large amounts of unvegetated bottom regardless of water levels. Herbicides
can further reduce plant populations in cultivated wetlands. Farmers use herbicides directly on cropped
wetlands, but also sometimes treat non-cropped wetlands to prevent introduced perennial grasses with
hydrophytic tendencies, such as Agropyron repens, from spreading to the uplands. Livestock grazing,
except when extremely intense, such as in heavily-trampled barnyards or feedlots, seldom creates
unvegetated bottoms.
Percent open water naturally increased in all zones as water was replenished after the drought
of 1992 and preceding years. The expectation that differences in watershed land use would result in
greater amounts of open water in sample wetland basins in good-condition watersheds held for all
zones during the relatively dry year of 1992, but in the following relatively wet year the differences were
less obvious, especially in zones of lesser water permanence. Open water was actually much higher in
wet-meadow zones in basins in poor-condition watersheds than those in good-condition watersheds in
1993.1 attribute this to the flooding of bare tilled soils created by cultivation.
107
-------
In summary, in poor-condition watersheds, the hydrology of prairie wetlands combines with a
variety of agricultural practices to create unnatural, coarse-grained patterns in wetland vegetation or
basins devoid of vegetation. Open water or unvegetated areas can lie adjacent to areas with greater
amounts of litter than are found where grazing is the predominant land use.
The effects of wetland phase were not tested because of sparse data, but in the dry year of
1992, communities in the drawdown phase were common. Those in the natural drawdown phase found
in wet-meadow and shallow-marsh zones of sample wetland basins in good-condition watersheds had
greater amounts of standing dead vegetation and less unvegetated bottom than those in these zones in
the cropland drawdown phase of basins in poor-condition watersheds.
I expected amounts of standing dead vegetation and litter in the normal emergent phase to be
greater in sample wetland basins in poor-condition than in good-condition watersheds because the
basins in the former likely would not be grazed and would often be idle, but no differences were
obvious. The expectation that there would be more open water in wetlands in good-condition rather
than poor-condition watersheds was based on commonly observed land use of the basins. Wetlands in
good-condition (grassland) watersheds tend to be grazed or mowed. This opens up dense stands or
emergents and often results in areas of greater amounts of submerged hydrophytes. Wetlands in poor-
condition watersheds tend to lie idle or support stands (often dense) of seeded crops intermixed with
annual weeds. While it seems true that tilled landscapes provide more runoff to the basins, it is the idle
basins that tend to choke up with emergent vegetation, especially the inner shallow-marsh and deep-
marsh zones that can only be farmed during dry years. Silt from farming operations seems to contribute
to the establishment of dense stands of emergents in these basins. Cultivation of basins in the poor-
condition watersheds likely reduced amounts of standing dead vegetation and litter to about the same
degree as livestock grazing did in basins in good-condition watersheds. As expected, deep-marsh
zones in the normal emergent phase generally had highest values of standing dead vegetation and
litter, but the values did not differ significantly between poor- and good-condition watersheds.
Communities in the cropland drawdown or cropland tillage phases of wet-meadow and shallow-
marsh zones in sample wetland basins in poor-condition watersheds had large amounts of unvegetated
bottom during the dry year as well as the wetter year of 1993. These phases directly reflect the effects
of intensive tillage. In the region, basins in the cropland tillage phase are often totally devoid of
vegetation, especially when these basins lie in fields undergoing summer fallow.
Plant species richness was lower in wet-meadow zones in sample wetland basins in
poor-condition watersheds. Lower species richness seemed directly related to the replacement of
108
-------
normal emergent and natural drawdown phases by cropland tillage and cropland drawdown phases. An
obvious pattern is the replacement of native perennials with introduced perennials, native and
introduced annuals, cultivated crop plants, and field weeds. Species richness in these basins is
probably further reduced by herbicides. Higher mean taxa richness in the normal emergent phase of
wet-meadow zones in poor-condition watersheds could reflect several phenomena including loss of
species through long-term grazing or haying of wet-meadow zones in the good-condition watersheds
and invasion of species into wet-meadows in poor-condition watersheds subject to a variety of current
and past disturbances. Relatively high mean taxa richness in communities in the natural drawdown
phase during 1992 could reflect the occurrence of normal plants of this zone combined with upland
plants that invaded during the preceding several years of drought. Very low mean taxa richness in the
cropland tillage phase of wet-meadow zones in poor-condition watersheds undoubtedly reflects the
replacement of normal emergent and natural drawdown species by cultivated crop monotypes and field
weeds that are usually subjected to treatment with herbicides.
As previously mentioned, the value of the abundance of submergents, the ratio of emergent
cover to open water in deep-marsh zones of semipermanent wetlands, and the abundance of
metaphytic or planktonic algae in open water areas in deep-marsh zones of semipermanent wetlands
could not be tested as EMAP indicators. No open water areas occurred in the deep-marsh zones of the
semipermanent wetlands selected for study in poor-condition watersheds. In any case, sample sizes of
semipermanent wetlands were too small in both watershed types to detect any differences in these
variables.
I conclude that several of the indicators we measured, especially amounts of unvegetated
bottom and plant species richness, successfully discriminated between wetlands in good- and poor-
condition watersheds. Intensively tilled prairie wetlands with large amounts of unvegetated bottom show
poor use by aquatic or marsh birds. For example, in the 1960's, these wetlands comprised about one-
fourth of the total area of basin wetlands in the North Dakota portion of the PPR. During this period,
only 4.4 percent of the ducks (Kantrud and Stewart 1977) and less than 0.5 percent of the other birds
(Kantrud and Stewart 1984) were observed on these basins. Nevertheless, there is a need for
additional indicators of the general environmental condition of wetlands. Most valuable would be
indicators that could be photographed or otherwise remotely sensed. A set of ideal indicators could
detect the absence of stressors as well as the presence of structures or functions of known value to
major groups of organisms.
109
-------
6.6 RECOMMENDATIONS FOR FUTURE EMAP STUDIES
1 suggest that measures of the amounts of unvegetated bottom and the presence of seeded
crops or cultivated soil in wet-meadow zones may be the best indicators of poor environmental
condition in prairie wetlands. Wet meadow zones are present in nearly all prairie wetlands. Unvegetated
bottoms, patterns in the soil left by various types of cultivation equipment, and the presence of seeded
crops could easily be interpreted from good quality aerial photographs taken at any time during the
growing season. Sampling time would be greatly reduced and landowner contacts would be
unnecessary because direct access to the basins would not be required.
Taxa richness was higher in communities in wet-meadow zones in good-condition watersheds
regardless of adjustments for community size, but this was not true for shallow-marsh and deep-marsh
zones. That the only significant correlation between taxa richness and community area was for
communities of deep-marsh zone within good-condition watersheds suggests that future EMAP studies
of taxa richness should concentrate on wet-meadow zones. However, use of species richness, and
subsequent consideration of species composition, as indicators of environmental condition for the next
phase of EMAP would require many landowner contacts because direct access to the basins by a
competent botanist would be necessary. Time expenditures per sampled basin would be much larger
than measures of environmental condition obtained remotely. Thus measures of species richness would
greatly reduce the number of basins that could be sampled per unit of effort.
Although untested during the Prairie Pothole Pilot Study, the abundance of submergent
vascular plants, cover/open water ratios, and the abundance of algae in open water areas in deep-
marsh zones of semipermanent wetlands may still be potentially useful indicators of environmental
condition of wetlands. The indicator value of semipermanent wetlands is limited, however, because they
compose only a relatively small proportion, perhaps 10-15% (Kantrud and Stewart 1984), of the basins
in any sampled land area in the region.
Future EMAP research could perhaps test a ranking system based on landscape-level
indicators for all undrained basins in a large geographical area. Watersheds could be ranked by the
degree of stress placed on the wetland basins by the most common land use practices on the adjacent
uplands. The presence of certain structures (open shallow water; turf of perennial plants) and functions
(maintenance of runoff volume; sediment retention) and the absence of certain stressors (excessive
herbage removal; mechanical disturbance) and problems (siltation; artificial drainage) could be noted for
110
-------
the outermost zone of the wetlands. Rank scores could be combined for uplands and wetlands to
create an index of environmental condition of wetlands.
The ranking system could be based on the premise that wetlands in the region had watersheds
where soils and vegetation had evolved primarily under the ecological influences of herbivory and fire
and were otherwise essentially undisturbed. Prior to agricultural disturbance, xeric grasses dominated
upland watersheds in the region; taller mesic grasses dominated low-prairie sites; and wetlands were
bordered by palustrine emergent vegetation. The ranking system would recognize that, because of
inherent natural fertility and ease of human access and occupancy, native vegetation types were the
focal points of disturbance by European agricultural and developmental practices and that these
practices and associated problems with high human population densities impact the condition of
wetlands to various degrees. The degree of alteration could be ranked numerically for uplands
(watersheds), low-prairie (here considered an ecotone or the major buffer between upland and
wetland), and the palustrine wet-meadow zone that forms the outer boundary of prairie wetlands.
Land units could be ranked through interpretation of aerial photographs or videographs, with no
need for ground surveys or landowner contact. Alternately, wetlands and their watersheds could be
ranked from roadside transects if only those wetlands clearly and wholly visible from the road and not
altered by road construction were surveyed. Whatever method is used, low rank scores could indicate
poor environmental condition of wetlands. Such a ranking system could use predetermined values to
score environmental condition of uplands, low-prairie, and wet-meadow which could be given
increasingly greater weights. Scores could be summed for an overall score for the wetland under
consideration. In hypothetical examples of this method (Tables 6-22 and 6-23), the maximum score is
21.0 for a grazed or burned temporary wetland that has not been cultivated, is surrounded by a grazed
or burned low-prairie zone that also has not been cultivated lying in a watershed comprised almost
exclusively of burned or grazed native grassland. Note that this ranking system could rank the
environmental condition of more complex wetlands (those with seasonally-, semipermanently- or
permanently-flooded, intermittently exposed, and saturated) under the assumption that intensity of
human land use always diminishes with increased period of flooding. Overall score for a larger
geographical area could be the mean rank for all wetlands in the area.
Ranking systems would also benefit from greater detail on past land use and intensity of land
use as shown by an example for wet-meadow zones (Table 6-24). These ranks could easily be
assigned to many wetlands by experienced observers during roadside surveys.
111
-------
Table 6-22. Hypothetical environmental condition scores for upland.
N>
Percent
Percent annually
developed tilled
land cropland
75-100 bulk of
remainder
75-100
75-100
75-100
75-100
75-100
75-100
75-100
75-100
Combinations
Percent Percent
hayed or grazed idle seeded
seeded cropland cropland
bulk of
remainder
bulk of
remainder
bulk of
remainder
bulk of
remainder
75-100 bulk of
remainder
75-100
75-100
75-100
75-100
of proportions of
Percent
hayed or idle
native
grassland
bulk of
remainder
bulk of
remainder
bulk of
remainder
bulk of
remainder
75-100
watershed land use
Percent
grazed or burned
native
grassland
bulk of
remainder
bulk of
remainder
bulk of
remainder
bulk of
remainder
bulk of
remainder
75-100
Score
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
-------
Table 6-23. Hypothetical environmental condition scores for low-prairie and wet-meadow zones of
prairie wetlands.
Predominant land use of zone
Low-prairie zone Score
Developed, drained, or silted in 0
Annually-tilled cropland or fallow i
Hayed or grazed, seeded to domestic grasses or legumes 2
Idle, seeded to domestic grasses or legumes 3
Idle native grassland 4
Hayed native grassland 5
Grazed or burned native grassland 6
Wet-meadow zone
Developed, drained, or silted in o
Annually-tilled cropland or fallow 2
Hayed or grazed, seeded to domestic grasses 4
Idle, seeded to domestic grasses 6
Idle native grassland 8
Hayed native grassland 10
Grazed or burned native grassland 12
The presence of recognized functions and absence of recognized stressors could also be
incorporated in a ranking system for wet meadows (Table 6-25).
The ranking system would recognize that current water conditions and land use practices have
drastic effects on wetland structure and function and resulting values. For example, flooded grassland
in grazed wet meadow zones produces large amounts of invertebrates compared to flooded bare soil.
During drought, the dry grassland of grazed wet meadows provides good nesting cover, whereas the
dry bare soil, crops, or crop residues of cultivated wet meadows is poor nesting cover. Such ranking
systems could be improved if other indicators of environmental degradation, such as partial drains,
upland gullies terminating in silt deltas in the wetland, or use of wetlands for feedlots or landfill sites
could be detected on the photographs.
Additional useful information could be obtained if multiple sets of photographs were available,
for instance from exposures taken during spring, summer, and fall. The ranking system could then be
refined to detect multiple stresses, such as fall plowing of the watersheds and basins (detectable in
spring), spring tillage and seeding of watersheds and basins (detectable in summer), and summer or
fall haying, grazing, burning, or recultivation of basins after harvest (detectable in fall). Lack of large
113
-------
Table 6-24. Example of an expanded ranking system for environmental condition of wet-meadow
zones.
Land use
Subtype Hypothetical
Intensity condition score
Cropping
Small grains 2
Row crops l
Fallowing
Mechanical 0
Chemical l
Grazing
Native vegetation
Heavy 6
Moderate 10
Light 8
Seeded or ruderal vegetation
Heavy 4
Moderate 6
Light 5
Mowing
Native vegetation 6
Seeded or ruderal vegetation 4
Burning
Native vegetation 9
Seeded or ruderal vegetation 4
Idling
Native vegetation
Seeded or ruderal vegetation
5
3
114
-------
Table 6-25. Land use of wet meadow zones of prairie wetlands as related to environmental condition
and water levels as indicated by critical structures, functions, and stressors".
Cultivated-high water levels
Structures Present- 1. Open shallow water
2. Smooth, bare bottom
3. Windrowed crop residue
4. Cropland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
6. Invertebrate production
7. Waterbird production
8. Habitat for migrant waterbirds
Stressor Absent- 1. Excessive herbage removal
2. Mechanical disturbance
3. Artificial drainage
"Score"--8 functions present +3 stressors absent=11
Cultivated-low water levels
Structures Present- 1. Rough bare soil
2. Crops or crop residue
3. Cropland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
6. Crop or forage production
Stressors Absent- 1. Artificial drainage
"Score"-6 functions present +1 stressor absent=7
Grazed-high water levels
Structures Present- 1. Open shallow water
2. Submerged turf of perennial hydrophytes
3. Grassland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
115
-------
Table 6-25. (continued)
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
6. Invertebrate production
7. Waterbird production
8. Habitat for migrant waterbirds
9. Winter wildlife cover
10. Crop or forage production
Stressors/Problems Absent- 1. Siltation
2. Tillage
3. Mechanical herbage removal
4. Pesticide use
5. Artificial drainage
"Score"-10 functions present +5 stressors absent=15
Grazed-low water levels
Structures Present- 1. Turf of perennial hydrophytes
2. Grassland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
6. Waterbird production
7. Winter wildlife cover
8. Crop or forage production
Stressors/Problems Absent- 1. Siltation
2. Tillage
3. Mechanical herbage removal
4. Pesticide use
5. Artificial drainage
"Score"--8 functions present +5 stressors absent=13
Mowed-high water levels
Structures Present- 1. Open shallow water
2. Submerged turf of perennial hydrophytes
3. Grassland or cropland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
116
-------
Table 6-25. (continued)
6. Invertebrate production
7. Waterbird production
8. Habitat for migrant waterbirds
9. Crop or forage production
Stressors/Problems Absent- 1. Siltation
2. Tillage
3. Pesticide use
4. Artificial drainage
"Score"-9 functions present +4 stressors absent=13
Mowed-low water levels
Structures Present- 1. Clipped turf of perennial hydrophytes
2. Grassland or cropland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
6. Waterbird production
7. Crop or forage production
Stressors/Problems Absent- 1. Siltation
2. Tillage
3. Pesticide use
4. Artificial drainage
"Score"-7 functions present +4 stressors absent=11
Idle-high water levels
Structures Present- 1. Tall, wet turf of perennial hydrophytes
2. Grassland or cropland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
6. Invertebrate production
7. Waterbird production
8. Winter wildlife cover
117
-------
Table 6-25. (continued)
Stressors/Problems Absent- 1. Siltation
2. Tillage
3. Pesticide use
4. Artificial drainage
"Score"--8 functions present +4 stressors absent=12
Idle-low water levels
Structures Present- 1. Rank turf of perennial hydrophytes
2. Grassland or cropland watershed
Functions Present- 1. Maintenance of runoff volume
2. Maintenance of runoff timing
3. Groundwater recharge
4. Sediment retention
5. Nitrate removal
6. Waterbird production
7. Winter wildlife cover
Stressors/Problems Absent- 1. Siltation
2. Tillage
3. Pesticide use
4. Artificial drainage
"Score"--? functions present +4 stressors absent=11
'The four stressors and nine functions are those identified as most important (score 4 or higher) in prairie wetlands by a panel of
experts as reported in Adamus, P.P. 1992. A process for regional assessment of wetland risk. U.S. Environmental Protection
Agency, EPA/600/R-92/249. The functions relating to maintenance of runoff, groundwater recharge, and nitrate removal have
been maintained for high water levels under the assumption that wet-meadows are nearly always dry prior to the following spring
snowmelt period. A tenth function, crop or forage production, has been added because it is the major economic use of wetlands in
the prairie region. The stressor pesticide use, although receiving a score of only 2 by the panel of experts, has been maintained
because of the possible reduction of hydrophytes caused by atrazine-type pesticides in the region. Wetlands are scored by
summing the number of functions present and number of stressors absent. The model assumes normal precipitation patterns for
the prairie region and that the presence of stressors is negatively correlated with wetland water levels. The model also assumes
weather conditions such that standard agricultural practices progress normally throughout the growing season.
amounts of wetland vegetation in wet basins during summer, particularly in fields of row crops, may be
a good indicator of herbicide damage to the wetland plants. The U.S. Department of Agriculture also
maintains files of fields where restricted pesticides are used. Overgrazing of watersheds and their
included wetlands could also possibly be detected if photographs of nearby reference sites known to be
more conservatively grazed were available for comparative purposes. Such ranking systems could be
improved if other indicators of environmental degradation, such as partial drains, upland gullies
terminating in silt deltas in the wetland, or use of wetlands for feedlots or landfill sites could be detected
on the photographs.
118
-------
Section 7.0
SOILS AND SEDIMENTS AS INDICATORS OF AGRICULTURAL
IMPACTS ON NORTHERN PRAIRIE WETLANDS
John A. Freeland and Jim L. Richardson
North Dakota State University
Department of Soil Science
Fargo, North Dakota
7.1 INTRODUCTION
The northern prairie wetlands are, for the most part, the products of glaciation that ended less
than 13,000 years ago (Bluemle 1991). Landscapes pocked by these wetlands are characterized by
internal drainage. Watersheds are the surrounding drainage basins that contribute runoff, ground water
seepage, dissolved solids, and eroded sediments to the wetlands located at the bottom. Some
watersheds are hydrologically isolated, while others are hydrologically and geochemically connected to
others by runoff or groundwater flow. Land use practices, especially agriculture, may have a significant
impact on both the quality and quantity of materials that enter wetland communities.
In time, wetland habitat will be lost or seriously damaged due to one or a combination of the
following natural or anthropogenic processes: (1) contamination by toxic levels of salts and other
chemicals, (2) establishment of an integrated stream drainage system, (3) long-term drought associated
with subsequent lowering of water tables, or (4) filling-in by inorganic sediments and organic matter.
While PO3 is readily adsorbed, NO3 is more likely to be transported in soluble form. Sediments often
carry phosphate and nitrate fertilizers (Neely and Baker 1989), which may enhance plant growth and
hasten the accumulation of organic sediments within the wetland basin. Elevated phosphorous levels
can promote eutrophication in aquatic ecosystems and enhance the growth of blue-green algae
(Schindler 1977, Crumpton 1989). Algae can become a problem to the point of causing fishkills in
prairie pothole lakes (Kling 1975). Chemical sediments, especially salts, are common in the subhumid
and semiarid prairie potholes. Salts can severely limit the condition and productivity of the potholes
(Richardson and Arndt 1989). However, some wetlands in the PPR contain plants and animals well
adapted for normal salinity, but elevated salinity from certain land-use practices are problems.
Soils are an essential part of the wetland ecosystem, serving as both a reservoir of water and
nutrients and as a medium for biogeochemical processes. To conserve wetland ecosystems, we need
119
-------
to understand processes that threaten them, be able to measure those processes and develop
acceptable protective strategies. Our primary objective in this study was to evaluate,
within the constraints of the project, the extent to which soils reflect the impact of land use, and identify
contrasting soil indicators that distinguish good-condition from poor-condition wetlands (See Section 2
for condition definition). To assess and monitor the condition of these wetlands, one must have a
concept of what constitutes good-condition wetlands, and be able to measure the processes which lead
to their demise. We are proceeding with the assumption that good-condition wetlands are those closest
to being pristine, i.e. undisturbed by human intervention, and poor-condition wetlands are those most
disturbed by human practices, especially agricultural tillage and cropping.
7.2 OBJECTIVES
1. Determine if good- and poor-condition wetlands can be distinguished from each other.
2. Determine the quality and quantity of sediment entering wetlands seasonally as a result
of erosion.
3. Determine the long-term sedimentation rates in good- and poor-condition wetland
landscapes.
4. Identify and measure some key soil constituents that reflect land use impacts, and pose
a threat to the condition of wetland ecosystems.
7.3 METHODS
7.3.1 Quality and Quantity of Sediments
Sediment trapping devices (see Section 9-1) were installed in 35 sample wetland basins. The
samples from the traps were analyzed for invertebrate remains at the Northern Prairie Science Center
(NPSC) as described in Section 5.6.1. The frozen sediment slurries were then mailed to us at North
Dakota State University (NDSU). We stored the samples for about one week in a walk-in refrigerator
maintained at a constant 3 °C. While in the refrigerator, the samples melted and the sediment settled to
the bottoms of their plastic containers. After settling, the clear water was decanted off and the
remaining sediment samples were dried for 24 hours in a forced-air evaporating oven at 65 °C. Since
120
-------
many of the samples from individual traps were too sparse for analysis, we composited the sediments
from each wetland into a single sample. Analyses included
• organic matter content using the loss-on-ignition method (Schulte 1988)
• calcium carbonate (CaCO3) equivalent by the Williams (1948) method
• phosphorus by the sodium bicarbonate (NaHCO3) extraction method (Olsen et al. 1954,
Knudsen and Beegle 1988).
We had planned to perform particle size analysis on the sediment samples, however, sediment
amounts were often too small to conduct the analysis.
7.3.2 Long-term Sedimentation (Cottonwood Lake Study Area)
Cesium-137 (Cs-137) is a radioactive isotope introduced to the atmosphere in the 1950's by
way of atomic weapons testing. Maximum atmospheric levels of Cesium-137 were detected in 1954,
and the winter of 1963-64. Cesium is tightly adsorbed to sediment particles and serves as a marker for
estimating sedimentation rates since 1954.
Several studies (DeLaune et al. 1978) have used peak Cs-137 levels found in sediment profiles
to successfully establish time markers and interpret depositional histories. The dating of vertical
sediment accumulation using Cs-137 peaks in sediment profiles depends on a major assumption we
feel we can not make in the northern prairie pothole wetlands. To use the profile-peak method we must
assume a constant sedimentation rate (Ritchie et al. 1973). However, using a hypothetical example,
suppose fallout Cs-137 entered a wetland basin in 1954 and remained attached to upland soils for
many years of relative drought and idle land use. If the basin was later disturbed by cultivation, then
received heavy precipitation in 1962, causing the land to erode, sediment laden with Cs-137 would
enter the wetland basin in 1962. If a researcher using the profile-peak method (described above)
sampled those new wetland sediments, which were high in Cs-137, they would probably interpret those
sediments as having been deposited in 1954, the first peak year for Cs-137 fallout. This, of course,
would be an erroneous interpretation, since the sediments actually were deposited in 1962. Considering
the highly variable climate here and the alternating drawdown and emergent phases of northern prairie
wetlands, constant sedimentation rates seem highly unlikely. Additional problems with the profile-peak,
121
-------
in our opinion, exist. For example, recharge events following a severe drought in a dry wetland may
cause clay particles bearing Cs-137 to be leached to lower profile depths. Cattle and other animals
common to the northern prairie wetlands may distort Cs-137 horizons by disturbing and mixing wetland
sediments.
For these reasons, we used another method similar to that used by Soileau et al. (1990), and
DeJong et al. (1986) to estimate average annual soil erosion rates. The method does not assume
constant sedimentation rates and focuses on soil loss occurring from 1954, the year when Cs-137 was
introduced to the atmosphere by weapons testing. The soil loss rate is averaged over 39 years. It
makes no difference how sporadic were the periods of erosion and deposition. Although we focus on
erosion using this model, not sedimentation, we think it is safe to expect watershed erosion to be
directly related to wetland basin sedimentation. Furthermore, it seems reasonable to expect that any
future land management practices aimed at reducing wetland sedimentation will have to directly
address watershed erosion.
We used the following analog of Soileau et al. (1990) to estimate annual erosion in cultivated
and uncultivated wetland watersheds, where:
A = [(B-C)/B] * D/E
A = annual rate of soil erosion (metric tons/ha),
B = total Cs-137 activity (Bq/m2) in 0-15 cm cores of baseline flat, non-eroded site,
C = total Cs-137 activity (Bq/m2) in 0-15 cm soil of eroded side slopes.
D = soil mass in 0-15 cm depth core (Mg/M3) * 1 ha volume of soil (metric tons),
E = years elapsed between initial Cs-137 fallout and soil sampling (39
years for this study).
Because of the high cost of Cs-137 analyses, we tested this method on four distinct sites as a
preliminary investigation of using Cs-137 to analyze sedimentation.
To determine long-term sedimentation rates in wetlands surrounded by cultivated versus
uncultivated fields, we collected soil samples from side slopes of four wetland watersheds (P1, P7, T1,
and an unnamed wetland basin on private property designated C7 for this study)in May 1993 (Fig. 7-1).
P1 and P7 are semipermanent wetlands; T1 and C7 are seasonal wetlands (Stewart and Kantrud,
1971). P1 and T1 are surrounded by grassland, C7 and part of P7 are surrounded by cultivated fields.
122
-------
COTTONWOOD LAKE STUDY AREA
Adapted from Winter ana Carr (1980)
Figure 7-1. Map of the Cottonwood Lake study area.
We collected duplicate samples, 5 cm in diameter by 15 cm deep, using a slide-hammer coring device
from three different points, all on side slopes, surrounding each of the four basins. Three samples of
the same dimensions were collected from a flat, uneroded site northwest of P7 to use for control (factor
B in the soil loss equation).
Subsamples were oven-dried at 105 °C to determine hygroscopic moisture and soil bulk
density. We composited the samples from each basin into single samples. Approximately 700 g of soil
composite were placed in four Marinelli beakers, one for each wetland, and analyzed twice for Cs-137
activity using gamma counting equipment. Counting time was 16 hrs. The gamma ray sensor was a 1.5
in. X 1.5 in. (3.8 cm by 3.8 cm) ORTEC 905-2 Nal Scintillation Detector coupled to a Canberra Series
85 Multichannel Analyzer. Analytical software was the MAESTRO II Emulation Software Model A64-BI
Version 1.40 (EG&G ORTEC, 1991, Oak Ridge, TN).
123
-------
7.3.3 Key Soil Constituents
7.3.3.1 Cotton wood Lake Study Area
Soil samples and data were collected at CWLSA in early June 1992. We collected soil samples
and data from four wetlands and wetland watersheds, including P1, T1, P7, and C7 (Figure 7-1). Soil
samples were collected from wetland vegetation zones {Stewart and Kantrud, 1971) along radial
transects. To establish the transects, we measured salinity every 10 paces along the wet meadow zone
of each of the four wetlands using the GEONICS EM-38 (Geonics LTD, Mississauga, Ontario). Other
transects were placed at about 100 m intervals, as measured by pacing, along the wet meadow. P1,
the largest wetland had 10 transects, P7 had 6, and T1 and C7 each had four transects. T1 and C7 are
relatively small wetlands, and the transects had to be spaced closer together to have a minimum of four
transects per wetland.
We collected soil samples from profiles where the transects intersected the wetland vegetation
zones at four depth increments per profile (0-15 cm, 15-30 cm, 30-45 cm, and 45-60 cm). We placed
about 400 grams of each sample in plastic-lined bags and stored them in coolers until late afternoon of
each field day, when we returned to the Woodworth Field Station. Here the samples were spread out in
a garage to air-dry. Back at NDSU, the air-dried samples were sieved through a 2-mm screen. All soil
laboratory analyses were conducted on dry, sieved samples.
We classified soil profiles in the field using Keys to Soil Taxonomy (Soil Survey Staff 1975).
Watershed measurements consisted of extending the linear transects away from the wetland to the top
of the wetland watershed divide. The length of each upland transect was measured by pacing. We also
measured steepness with a pocket clinometer.
The specific soil characteristics that we tested for their use as potential condition indicators
included
• Soil Classification (Soil Survey Staff, 1975)
• Nitrate-Nitrogen by transnitration of salicylic acid (Vendrell and Zupacic 1990)
Sodium bicarbonate-extractable Phosphorus (Knudsen and Beegle, 1988)
124
-------
• Organic Matter by loss-on-ignition (LOI) (Schulte 1988)
• Soluble Salts by electrical conductivity (EC) (Dahnke and Whitney 1988)
• In Situ Salinity using a GEONICS EM38 electromagnetic induction meter
• Soil pH in 2:1 0.01 M CaClz solution-to-soil slurries (Eckert 1988)
• Particle Size using Particle Size Analysis (PSA) by the hydrometer method (Day 1965).
The nitrate, phosphorus, and organic matter tests were performed at the NDSU Fertility
Laboratory; soluble salts, pH and PSA were performed in the NDSU Soil Characterization Laboratory.
The sodium bicarbonate (NaHCO3)-extractable P is sometimes known as the "Olsen" test
(Olsen et al. 1954) after its originator. This is a relatively inexpensive procedure that yields values
which have correlated well with crop responses. This test is the one routinely used on agricultural soils
in North Dakota in order to make fertilizer recommendations. Since we are concerned with fertilizer
additions to wetlands, the Olsen test seems appropriate, since it is sensitive to common fertilizer
sources of P. Wolf et al. (1985) found fairly good coefficients of determination (0.71) between Olsen
test P and algal available P.
The LOI method (Schulte 1988) was developed as an alternative to the more complex and
time-consuming Walkley-Black (1934) carbon test, which is a wet-chemical procedure. The LOI method
correlates very strongly to the Walkley-Black test and, unlike Walkley-Black, requires no hazardous
chemicals. The LOI test includes first drying a sample to 105 °C, and recording its dry weight. Next the
sample is baked in a muffle furnace at 360 °C for 2 hours and weighed again. Although LOI tests are
often run at 450 °C or higher; in prairie regions, 360 °C should be the standard procedure. The weight
loss is due to the oxidation of organic matter (OM) in the soil.
The test for soluble salts by electrical conductivity (EC) was performed on 1:1 soil to distilled
water slurries. Twenty ml of water were added to 20 g of soil, stirred, and left to stand for 15 minutes
before measuring electrical conductivity with a Type 700 Conductivity Meter (Chemtrix Corp., Hillsboro,
OR). Soil pH was measured using an ORION Ion Analyzer Model 901 (Cambridge, MA).
125
-------
Particle size analysis was performed on samples collected from four depth increments: 0-15
cm, 15-30 cm, 30-45 cm, and 45-60 cm. Prior to analysis, samples were air-dried and sieved through a
2-mm screen. We used a hydrometer method similar to that described by Day (1965) with some
modifications intended to save time and allow for the analyses of greater numbers of samples. Our
method does not include digestion of organic matter, washing salts, or shaking overnight; it emphasizes
chemical instead of mechanical dispersion. The procedure was as follows:
1. Weigh 40 g of soil and place it in a hydrometer jar.
2. Add 100 ml of Calgon dispersant.
3. Add enough distilled water to the Calgon to make 1 liter.
4. Agitate the suspension with 30 up-and-down plunger cycles and let sit overnight.
5. In the morning, check temperatures, agitate again with 30 plunger cycles and begin
hydrometer readings.
6. Read hydrometer values at 1, 3, 10, 30, 90, 270, and 480 minutes.
7. Calculate sand, silt, and clay percentages using a LOTUS 123 spreadsheet in the Soil
Characterization Laboratory.
The procedure saves time on bottle washing since there is no overnight shaking in a flask or
drink mixer. Again, we performed no salt washing or OM digestion.
Multiresponse permutation procedure (MRPP) is a statistical method of analyzing ecologic data
which do not necessarily conform to assumptions of normal distribution and equal variances required
when using least squares analyses such as linear regression and analysis of variance (Biondini et al.
1988). Since some of the data generated by this study do not fit the normal distribution, use of MRPP
seemed appropriate.
MRPP is a method based on absolute Euclidean distances (Biondini et al. 1988). Distances are
calculated between all possible pairs of points in each group and averaged to calculate the group
distance value. Then, the group distance values are weighted according to the number of samples in
126
-------
each group and averaged together to calculate the delta value. If the groups are real, i.e., they form
separate groups in the data space, the delta value will be significantly low. If, however, the assigned
groups are not really different, alternative groupings, or permutations, may yield distance values smaller
than the values calculated for the chosen groupings. In this latter case, the P-values would be high,
since the chance of getting lower delta values would be relatively high.
The basic statistical data are in Tables 7-1 and 7-2. MRPP was applied to nitrate, P, and OM
data from the 0-15 cm depth samples collected from each transect. The values shown in Table 7-3 are
from wet meadow soils, or in cases where wet meadow was absent, shallow marsh soils. The samples
shown come from the 0-15 cm depth increment. These samples were chosen for primary analysis since
they are closest to the surrounding land use activities, i.e., they are on the edge of the wetland, on or
near the land surface. Samples were separated into two conditions, based on the land use adjacent to
and up-slope from the sample site. In P7, transects 2 and 6 were somewhat intermediate in their
adjacent land use. The immediately-surrounding land use is grassland for those two sites, but, we think,
they are sufficiently close to the cultivated fields that they warrant placement in the cultivated
(poor-condition) group. Seventeen samples fell into the grassland (good-condition) group, and 7 were
placed in the cropland (poor-condition group).
7.3.3.2 10.4-km2 Sample Plots
In 1992, we sampled 40 wetlands at the same sites where H. Kantrud took plant data (see
Section 6.3.1) and collected soil samples from vegetation quadrats placed in delineated plant
communities. In each community, soil profiles from five quadrats per community were classified, and
soil was collected from quadrats two and four for laboratory analysis. Soil profiles were dug, using a
"Dutch" auger, to a depth of about 75 cm, sufficiently deep to classify the soil and collect laboratory
samples. Two samples were collected from each profile, one from 0-15 cm depth and another from 15-
45 cm. Laboratory analyses included all those performed at the CWLSA wetlands except PSA, i.e.,
NO3' NaHCO3-extractable P, organic matter, EC and pH. As in the CWLSA study, we measured basin
size by pacing from the wet meadow to the basin divide along at least 4 transects. We also measured
in situ salinity along the wet meadow using the EM-38 in the same manner described above for the
CWLSA study.
In 1993, we sampled soils from 36 randomly selected wetlands according to wetland vegetation
zones from H. Kantrud's study (see Section 6). We sampled and classified soil profiles from 3 of the 5
127
-------
quadrats selected per community. As in 1992, we collected samples at 2 depths, 0-15 cm and
15-45 cm. Samples were bagged in plastic-lined paper bags and stored in coolers, as they were in
1992. Approximately 20 grams of each bagged sample were placed in aluminum cans, which we
opened in the evening of each field day and placed in a forced-air evaporating oven. We carried the
oven in our vehicle, setting it up in a motel room in the evening. Samples oven-dried at 65 °C overnight
at a motel. Prompt drying is necessary to prevent analytical errors due to potential nitrogen
transformation by microbes in the sample bags (Dahnke, 1988). The oven-dried samples were lightly
ground, sieved through a 2-mm screen and used for NO3", P and OM analyses. The bagged samples
were used for analysis of pH, EC, and PSA. We only ran PSA on the 0-15 cm depth samples. Soil pH
in 1993 was measured in distilled water, instead of 0.01 M calcium chloride (CaCI2) as was used in
1992. In the above laboratory tests, except particle size analysis, every 10th sample was replicated.
We used analysis of variance (ANOVA) techniques to assess the effects of wetland condition,
zone (deep marsh, shallow marsh, and wet meadow), depth, and year (1992, 1993) on the response
variables NO3, P, OM, EC, pH, sand, silt, and clay. The design was a strip-split-plot with repeated
measures. Each basin was assumed to be the independent whole-unit, with zone-depth and community
combination being the subunit. Because most watersheds were measured in both 1992 and 1993, year
served as the repeated measures factor. We used Fisher's protected least squares differences (LSD) to
isolate differences in least squares means following significant effects in the ANOVAs (Milliken and
Johnson 1984). All ANOVAs were done using the general linear model procedure (PROC GLM) of SAS
(SAS Inst. 1992). A 1n(y+1) transformation was done on all data except pH prior to analysis because
the data were skewed to the right (note: clay was only marginally skewed). Data were averaged across
quadrants (2 in 1992, 3 in 1993) prior to 1n(y+1) transforming. Sand, sift, and clay were only measured
at the 0-15 cm depth in year 1993. Statistical tests were considered significant at the 0.05 level, and
marginally significant at the 0.10 level.
7.3.4 Soil Oxidation-Reduction
We placed platinum electrodes in the three wetland zones (wet meadow, shallow marsh, and
deep marsh) of wetlands T1, P7, and C7. We wanted to monitor oxidation-reduction potential to see if
the hydric soil morphology observed in the sampled profiles at CWLSA corresponded to active
oxidation-reduction processes, or, alternatively, the hydric soil morphology was possibly relict, i.e.,
developed during some wet climatic episode in the distant past. The electrodes were to be monitored
during the frost-free months of September 1992 to November 1993.
128
-------
7.4 RESULTS
7.4.1 Seasonal Sedimentation
Results of chemical analyses of trapped sediments are shown in Table 7-1. In this table, CRP
land is considered good-condition because short-term sedimentation is expected to reflect the most
recent activities within the drainage basin. Average amount of sediment per trap and phosphorus
concentrations were nearly equal for the good- and poor-condition groups. Average organic matter in
the good-condition group (15.1 g/100 g) was significantly greater (F=7.14, df 1,29 P = .012), than in the
poor-condition group (9.4 g/100 g). Calcium carbonate (CaCO^ equivalent (CCE) is a measure of that
percentage of the sediment mass attributable to CaCO3. We did not have sufficient sample to perform
this test on many samples and, therefore, we are reluctant to draw any comparisons between condition
groups. In sediment analysis, we expected high variation because of the wide settlement rate due to
minor landscape and vegetation differences, creating, large settlement differences, both in amount and
quality of sediment. Sediment will always have problems in use.
7.4.2 Long-term Sedimentation
Estimations of soil loss for the four wetland basins P1, T1, P7, C7 and a flat, noneroded site
used for making comparisons are shown in Table 7-2. Using the formula from Soileau et al. (1990), we
calculated soil loss (tons/ha) for each basin. As expected, the cultivated basin C7 had the greatest soil
loss from its side slopes (35.6 metric tons/ha/year), and P7, the basin partially surrounded by cultivated
fields had the second highest soil loss (11.71 metric tons/ha/year). T1 soil loss was 4.56 metric
tons/ha/year. The P1 soil loss value (+16.13) indicates a net gain in Cs-137 and soil deposition
compared to the noneroded control site.
7.4.3 CWLSA Soil Characterization
Results of laboratory analyses for fertilizer nutrients NO3", P and OM are shown in Table 7-3.
Results from the MRPP (Table 7-4) show that P is the strongest contrasting variable separating
the good from the poor group, while the NO3 variable alone is the least significantly different.
129
-------
Table 7-1. Analysis of selected chemical constituents of trapped sediments, 34 sample wetlands, North and South Dakota, collected in 1993.
Abrev.: OM = organic matter, CCE = calcium carbonate equivalent, SD = standard deviation.
Plot
73
73
133
133
133
134
134
134
134
134
134
156
156
156
327
327
327
363
363
374
374
374
374
407
407
407
442
442
442
442
442
Wetland
29
86
370
380
386
140
158
165
270
406
432
22
26
42
72
117
147
22
58
65
100
225
272
67
109
168
93
260
261
281
295
Condition
Good
Good
Poor
Poor
Good
Poor
Poor
Poor
Poor
Poor
Poor
Good
Good
Good
Poor
Poor
Poor
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Poor
Poor
Poor
Good
Dry Density
(g/cm3) Pg/m3
0.39
0.48
0.76
0.39
0.39
0.36
0.78
0.55
0.54
0.82
0.45
0.62
0.68
0.39
0.74
0.70
0.57
0.47
0.36
0.50
0.32
0.38
0.72
0.65
0.43
0.83
0.53
0.88
0.86
0.66
0.28
69.6
35.2
55.4
111.0
139.0
38.8
115.6
93.8
54.6
46.8
36.1
39.1
45.5
89.4
75.8
47.7
54.0
14.6
23.5
60.4
32.5
50.3
13.7
54.2
36.6
43.8
60.4
56.2
64.0
64.0
51.2
Pg/Mg
178.5
72.7
73.2
285.9
358.0
106.9
147.8
169.5
102.0
57.3
79.4
63.2
66.9
230.3
101.9
68.3
94.7
31.0
65.6
121.9
101.6
132.8
19.1
83.8
84.7
53.0
113.3
64.2
74.6
97.0
184.0
Sediment
g/trap
0.87
1.04
3.53
0.95
0.27
0.90
6.06
1.58
0.88
2.13
0.65
6.43
7.89
0.34
2.56
4.75
4.32
9.31
0.45
4.20
1.40
5.98
1.45
4.17
0.43
9.63
1.37
6.12
5.23
2.62
0.69
P loading
rate
g/Mg
155.3
75.6
258.3
271.6
96.7
96.2
895.4
267.7
89.8
122.0
51.6
406.3
527.9
78.3
261.0
324.5
409.3
288.5
29.5
512.2
142.2
794.0
27.7
349.3
36.4
510.4
155.3
392.7
390.0
254.1
127.0
%OM
11.9
23.9
8.6
INS
INS
11.8
8.1
10.5
13.1
7.5
10.7
14.3
10.0
INS
9.1
11.4
12.5
11.4
18.6
26.4
25.3
24.1
3.7
13.0
21.0
6.1
11.6
5.7
5.9
7.7
23.6
% CCE
INS
0.0
0.0
INS
INS
INS
3.8
INS
1.8
0.6
INS
0.0
0.0
INS
3.6
0.0
0.0
0.3
INS
10.2
2.4
0.0
4.9
0.0
INS
10.0
INS
0.0
0.0
0.0
0.0
-------
Table 7-1. (Continued)
Plot
442
498
498
Mean
Mean
SD
SD
Dry Density
Wetland Condition (g/cm3) P g/m3
301 Good 0.41 70.1
146 Good 0.69 39.7
277 Good 0.60 41.2
Good
Poor
Good
Poor
Sediment
P g/Mg g/trap
169.8 0.79
57.6 B.92
68.2 3.22
3.4
3.0
3.4
2.0
P loading
rate
g/Mg
134.1
513.7
219.6
259.0
291.7
217.6
209.2
%OM
13.0
6.7
7.5
15.1
9.4
7.4
2.4
% CCE
0.0
0.0
0.0
-------
Table 7-2. Estimation of soil loss in four CWLSA wetlands plus a non-eroded control site using
Cs-137 analysis (Soileau et al. 1990). Bulk density values are mean values from three field
samples, Cs-137 activities are mean values of two laboratory runs. Each Cs-137 sample
was a composite of three basin subsamples. Gamma ray count time was 57600 seconds.
P1 T1
Bulk Density
(Mg/m3) 0.83 1.17
Cs-137
Activity Bq/m2 5.oi 2.99
Soil Loss (-)
or gain (+)
(metric tons +16.13 -4.56
P7 C7 Non-eroded"
0.89 1.39 1.20
2.20 1.11 3.33
-11.61 -35.64 0.00
Combining P with one or both of the other two variables does not improve the P-value, or seemingly
add to the separation of the two condition groups. Distances between phosphorus values in the
poor-condition group are about twice the distances in the good group, indicating unequal variances in
the two groups of data, one of the reasons for using MRPP.
7.4.4 Results of 1992 and 1993 EMAP Soil Characterization
Poor wetland condition included land recently placed into CRP, since the soil analyses probably
reflect long-term conditions in the wetland basin. Least squares (LS) means and mean comparisons
were made from log-transformed data Tables 7-5, 7-6, and 7-7. Log-transformation was necessary on
all but the pH variables, due to data distributions being skewed to the left. Back-transformed means are
also shown in those tables.
Nitrate varied significantly with year (F195=9.46; P=0.0027), and with depth (F, 48=13.559; P =
0.0006). The year effect implies that the differences between 1992 and 1993 are consistent for
condition, zones and depths (i.e., comparisons between years can be made by ignoring condition, zone
and depth). The depth effect implies that depth differences are consistent between condition, zones,
and years. Phosphorus varied significantly with year (F19S=5.02; P=0.0274), zone (F235=6.57; P=0.0038)
and marginally with condition and depth interaction (F, 40=2.88; P=0.0962). OM only varied significantly
with depth (F148=108.69; P=0.0001).
132
-------
Table 7-3. CWLSA Soil nutrient analysis of 0-15 cm samples from wetlands P1, T1, P7, and C7,
Cottonwood Lake Study Area, 1992. Abrev.: WM = wet meadow, SM = shallow marsh,
= nitrate. Units: OM = % mass, NO, and P g/m3.
Wetland
P1
P1
P1
P1
P1
P1
P1
P1
P1
P1
T1
T1
T1
T1
P7
P7
P7
P7
P7
P7
C7
C7
C7
C7
Transect
l
2
3
4
5
6
7
8
9
10
1
2
3
4
1
2
3
4
5
6
1
2
3
4
Condition
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Good
Poor
Poor
Good
Good
Good
Poor
Poor
Poor
Poor
Poor
Zone
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
SM
WM
SM
SM
WM
WM
WM
WM
WM
WM
WM
WM
OM
4.3
4.2
2.4
2.6
1.9
2.2
5.1
3.2
2.6
3.1
4.9
6.0
8.9
6.4
9.5
9.8
11.7
6.3
6.9
7.9
9.9
2.2
9.4
9.6
N03
8.1
9.9
6.2
6.3
5.0
5.0
13.7
6.3
3.8
5.6
6.3
5.0
13.7
11.2
13.8
17.4
11.9
8.1
11.2
8.2
13.7
1.9
6.3
11.3
P
6.9
7.4
5.0
3.8
4.4
5.6
7.5
5.0
6.3
6.9
8.1
6.2
34.3
6.9
33.8
26.1
11.3
7.5
7.5
8.8
35.0
11.3
20.7
35.2
133
-------
Table 7-4. MRPP statistical analysis of soil nutrient data, CWLSA, 0-15 cm soil depth, 1992.
Variable
Phosphorus
Nitrate and
Organic Matter
Organic Matter
Phosphorus and
Organic Matter
Condition
Good
Poor
Good
Poor
Good
Poor
Distance
Observed Delta
8.01
15.74
2.93
2.58
6.38
13.88
10.27
2.82
8.57
P-Value
Phosphorus
Phosphorus
and Nitrate
Nitrate
Good
Poor
Good
Poor
Good
Poor
4.96
13.3
7.07
15.30
3.70
6.38
7.39
9.47
4.48
0.00047
0.00071
0.26474
0.00071
0.00319
0.00049
EC varied significantly with zone (F235=3.42; P=0.044) and with depth (F148=3.93; P=0.0531).
pH varied with year (Ft 9S=3.66; P = 0.0588), zone (F2i3S=3.93; P=0.0289), and depth (F148=13.59; P =
0.0006). Sand (F233=21.19; P=O.OQ01), silt (F233=5.42; P=0.0092) and clay (F233=7.79; P=0.0017) varied
significantly only with zone. All other effects and interactions were nonsignificant.
Tables 7-5 and 7-6 show Fisher's protected LSD tests and LS means for log-transformed NO3
and P data. The letters a, b, and c located next to the LS Mean values indicate whether or not the
means are significantly different. Table 7-7 shows Fisher's protected LSD tests for sand, sift, and clay.
Values followed by another value with a common letter, for example shadow marsh and wet meadow
silt in Tabte 7-7, are not significantly different. Significant differences by year and depth increment
occurred for nitrate values. The 7.5 and 30 cm depths shown in the tables indicate the midpoint of the
0-15 cm and 15-45 cm depth increments we sampled in the field. Significant differences of P levels
exist by year, by zone, and by condition*depth interaction. Higher P concentrations occurred in 1993,
and concentrations were highest in the deep marsh zone. The condition*depth interaction indicates the
P concentrations in the 0-15 cm depth soils in the poor-condition wetlands were significantly higher than
in the good-condition soils of the same depth increment, independent of zone or year.
134
-------
OM mean concentrations (Table 7-6) were significantly higher in surface soils compared to
subsoils, independent of other interactions. This is usually the case in soils, where decaying surface
vegetation and microbial activity leave higher concentrations of OM near the surface. EC (Table 7-6)
was significantly higher in deep marsh zones (DM) than in either shallow marsh (SM) or wet meadow
(WM) zones. Wet meadow and shallow marsh salinity (EC) were not significantly different. The Fisher's
test also found significant differences in pH values by year, by zone, and by depth (Table 7-6). Clay
content was highest in deep marsh zones (Table 7-7).
7.4.5 CWLSA Soil Oxidation-Reduction Potential
All but one (0.00 being neutral) of the soil oxidation-reduction measurements from September
and October 1992 are positive, indicating relative oxidizing conditions (Table 7-8). May and June 1993
values are mixed, but mostly negative, indicating relative reducing conditions. Interior zones of T7 and
P7 flooded deep enough to prevent us from monitoring soil oxidation-reduction. Wetland P1 was not
monitored.
7.5 EVALUATION
7.5.1 Seasonal Sedimentation
Results of the chemical analyses of the trapped sediments did not reveal significant differences
in P inputs occurring between the sampled good- and poor-condition wetlands. Organic matter makes
up a larger proportion of the sediments in good-condition wetlands than in the poor-condition wetlands
(8.3 g/100 g). A greater proportion of the poor-condition sediment is mineral material. This is a
potentially important indicator reflecting higher rates of erosion and sedimentation in the poor-condition
wetlands. While organic matter will, for the most part, be decomposed to biologically recyclable
nutrients and gases, the inorganic sediment will remain mostly inert and, given time, fill in the wetland.
7.5.2 Long-term sedimentation
The Cs-137 study addressed the problem of sedimentation indirectly by examining soil loss
from the wetland watershed side slopes. The C7 wetland had the highest rate of soil loss. However the
value for P1 indicates Cs-137 and soil deposition on its side slopes.
135
-------
Table 7-5. Least squares means for nitrate and phosphorus, EMAP sample wetlands, 1992-93.
Variable
Nitrate
Phosphorus
Effect
YEAR 1992
1993
Pooled MSE =
DEPTH 7.5
(cm) 30
Pooled MSE =
YEAR 1992
1993
Pooled MSE =
ZONE DM
SM
WM
Pooled MSE =
CONDITION
High
Low
Pooled MSE =
LS Mean8
ln(Y+1)
1.88 b
1.42 a
0.2508
1.80 b
1.49 a
0.1563
3.26 a
3.44 b
0.1348
3.62 (40) c
3.34 (86)
3.08 (159) a
0.4759
DEPTH (cm)
7.5 30
3.36 (78)a 3.09 (78)a
3.75 (65)b 3.19 (64)b
0.0926
Back-transformed
LS Mean
7.55
5.14
7.05
5.44
27.05
32.19
38.34
29.22
22.76
DEPTH (cm)
7.5 30
29.79 2.98
43.52 25.29
'Within a column, LS Means followed by a common letter are not significantly different at the 0.05 level using Fisher's protected LSD value.
-------
Table 7-6. Least squares means for log transformed percent organic matter (OM) and electrical conductivity (EC), EMAP sample wetlands, 1992-93.
pH data were not log transformed.
CO
Variable
%OM
EC
(micromhos)
PH
Effect
DEPTH
(cm)
ZONE
DEPTH
(cm)
ZONE
YEAR
DEPTH
(cm)
7.5
30
Pooled MSE =
DM
SM
WM
Pooled MSE =
7.5
30
Pooled MSE =
DM
SM
WM
Pooled MSE =
1992
1993
Pooled MSE =
7.5
30
Pooled MSE =
LS Mean"
2.10 (143) b
1.64 (142) a
0.0662
6.75 (40) b
6.33 (86) a.
6.19 (159) a
0.6464
6.38 (143)
6.46 (142)
0.0681
7.15 (40) b
6.94 (86) a
7.12 (159) b
0.1476
6.96 (143) a
7.18 (142) b
0.1924
6.96 (143) a
7.18 (142) b
0.1228
Back-transformed
LS Means
9.17
6.16
855.06
562.16
488.85
590.93
640.06
"Within a column, LS Means followed by a common letter are not significantly different at the 0.05 level using Fisher's protected LSD value.
-------
Table 7-7. Least squares means for log-transferred percent sand, sift, and clay; EMAP sample wetlands,
1993.
Variable
Effect
LS Means 1n(Y+1)a
Back Transformed
LS Mean
Sand
Silt
Clay
ZONE DM
SM
WM
ZONE DM
SM
WM
ZONE DM
SM
WM
2
3
3
4
3
3
3
2
2
.60
.22
.45
.20
.92
.85
.32
.95
.81
(11)
(22)
(39)
(11)
(22)
(39)
(ID
(22)
(39)
a
b
c
b
a
a
b
a
a
14
26
32
67
51
47
28
20
17
.46
.02
.5
.69
.40
.99
.66
.11
.61
"Within a column, LS Means followed by a common letter are not significantly different at the 0.05 level using Fisher's protected
LSD value.
Table 7-8. Soil oxidation-reduction potential measurements from 3 CWLSA wetlands. September,
1992-June, 1993.
Date
Wetland
Zone
Rep
Depth (cm)
(mVolts)
09-10-92 T1
10-07-92 T1
05-13-93 T1
06-17-93 T1
09-10-92 T7
T7
10-07-92 T7
WM
SM
WM
SM
WM
SM
WM
SM
WM
SM
WM
1
1
1
1
1
1
1
1
1
1
1
1
1
l
l
1
1
2
1
2
1
2
1
2
1
2
1
45
15
45
15
45
15
45
15
45
15
45
15
45
15
45
15
45
45
15
15
45
45
15
15
45
45
15
+ 235
+196
+210
+240
+311
+301
+318
0.00
+148
+200
-428
-335
-107
-170
-69
-46
+ 350
+ 374
+ 365
+ 337
+ 282
+ 340
+ 310
+ 313
+ 335
+ 380
+442
138
-------
Table 7-8. (Continued)
Date
05-13-93
06-17-93
09-10-92
10-07-92
05-13-93
~
06-17-93
Wetland
T7
T7
T7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
Zone
SM
WM
SM
WM
SM
WM
SM
DM
WM
SM
WM
SM
DM
WM
SM
Rep
2
l
2
l
2
1
2
1
2
FLOODED
1
2
1
2
FLOODED
1
2
1
2
1
2
1
2
1
1
1
2
1
2
1
2
1
2
2
1
1
2
1
2
1
2
FLOODED
1
2
1
2
1
2
1
2
Depth (cm)
15
45
45
15
15
45
45
15
15
45
45
15
15
45
45
15
15
45
45
15
15
45
15
45
45
15
15
45
45
15
15
45
45
15
15
45
45
15
15
45
45
15
15
45
45
15
15
(mVolts)
+418
+345
+393
+410
+420
+10
-265
-254
-200
-159
-250
-152
-208
+ 270
+280
+207
+154
+253
+246
+ 248
+ 245
+250
+ 245
+290
+272
+254
+264
+278
+288
+314
+317
+201
+ 195
+217
+ 180
-144
+107
-302
-272
-205
-120
+214
+260
-504
-510
-460
-480
The higher Cs-137 values on P1 slopes could be due to snowcatch on the grass-covered side slopes,
with the snow containing Cs-137 fallout. Snow particles nucleate around dust particles in the air and
thus would be added to soil upon melting. Another possibility is that explaining the apparent soil
139
-------
deposition on P1 side slopes of P1 is soil loss from the flat, "noneroded" site. We see no evidence,
however, to support this. Another possibility, slow soil creep on the P1 side slopes may have moved
Cs-137-attached soil from higher on the hill slope down to and on top of the mid-slope position where
we collected our samples. Finally, the problem of apparent soil deposition of the P1 side slope could be
explained by inadequate sample size.
7.5.3 Soil Characterization: CWLSA and EMAP Studies
Soil NO3 was significantly lower in 1993 than in 1992. At least three factors might explain this
contrast: denitrification, leaching, and change in sample handling procedures. Summer 1992 marked
the end of a drought in the northern plains. Relatively high precipitation during the second half of 1992
and first half of 1993 refilled previously dried wetlands. Denitrification occurs as a result of chemical
reduction usually associated with saturated, anaerobic environments. Nitrate, an oxidized nitrogen (N)
compound is converted to more reduced compounds including ammonia (NH3), nitrous oxide (N2O), and
nitrogen (NJ, all three of which are gaseous and return to the atmosphere. This natural process may
have produced lower NO3 levels in 1993. NO3 is also highly soluble and may have leached to deeper
soil depths in 1993. Finally, the lower NO3 values in 1993 may have been due to a procedural change.
In 1993, we dried soil samples overnight in an evaporating oven. This relatively fast drying was done to
help eliminate possible oxidation of formerly reduced N compounds upon exposure to air. In 1992,
samples had longer exposure to air that could have caused NO3 values to be elevated.
NO3 varied significantly with depth, the higher concentrations being in the 0-15 cm samples.
This is most likely due to the greatest portion of the total soil nitrogen pool's association with soil
organic matter. Decomposition and oxidation of organic nitrogen produces higher NO3 concentrations in
the topsoil.
From the studies at the CWLSA and EMAP sample wetlands, P was the strongest indicator
showing the apparent impact of cultivation on wetland nutrient concentrations. Phosphorus, unlike
nitrogen, has no stable gaseous forms and, once in the wetland, tends to remain there. Although
nitrogen fertilizer is commonly applied to cultivated fields, denitrification under reducing conditions
results in reduced gaseous forms of nitrogen escaping to the atmosphere.
Considerable evidence exists linking P loading to runoff and soil erosion. Andraski et al. (1985)
compared P losses in runoff from four different tillage schemes including conventional till, chisel plow,
140
-------
till-plant, and no-till. They measured total P, dissolved molybdate-reactive P (DMRP) and algal-available
P (AAP). It is important to note here that there are differences in soil P forms. Some are more available
to plants than others. The DMRP and AAP are of higher ecologic importance than total phosphorus,
since much of the total P is tied up in relatively insoluble minerals. Conservation tillage greatly reduced
total P and sediment runoff. DMRP runoff from conservation tilled plots was lower than or equal to
runoff from the conventionally, tilled plots. The AAP runoff comparisons, perhaps the best indicator of P
pollution, were highest for the conventionally tilled plot and the till-plant plot, and lowest in the no-till
plot. Andraski et al. (1985) stated that the algal-available P includes the dissolved (DMRP) phosphorus
and about 20% of the total P. AAP is made up of about 30% DMRP, and 70% is inorganic, easily
desorbed or easily dissolved particulate P associated with the sediment. The important point here is
that the algal available P is mostly particulate phosphorus.
Since soil P can take many forms which might play important ecological roles, we need to know
which ones we are measuring when performing a soil test. Wolf et al. (1985) examined different soil
tests, including the Bray-l, Olsen, and Mehlich I methods to see how well they corresponded to
equilibrium-dissolved-P concentration, "labile" P, and AAP. For north-central soils, the Olsen NaHCO3
soil test P related significantly to the AAP.
Organic matter varied significantly with depth, as was expected. Under natural soil conditions,
organic matter decreases with depth. Under extreme cases of erosion and sedimentation, however,
where mineral soil is deposited on top of more organic-rich topsoil, organic matter increases with depth.
Such a profile may be a good indicator of disturbance.
Although pH and EC do not appear to be valid indicators of wetland condition, they do provide
an interesting reflection of climatic changes in the region. In our study pH was shown to be significantly
.different between years, but this fact, we are quite sure, is due to a laboratory procedural change in the
way we measured pH. In 1992, we measured pH in 0.01 M CaCI2, which forces H* ions off soil
exchange sites and, subsequently, lowers the measured pH value. In 1993, we measured pH in distilled
water. By doing so, we could use the same sample prepared for the EC analysis to measure pH. This
saved time and reduced the total amount of soil needed to carry out the full set of lab analyses. Our
data tell us the soil pH measured in CaCI2 is 0.1 to 0.4 units lower than the same soil measured in
distilled water.
The soil texture analyses showed significant differences between zones, as expected. Coarser
soil textures are found along the wetland edges, finer textures in the wetland interiors. Since less
141
-------
energy is required to transport silt or clay, wave action along wetland edges sorts fine particles and
relocates them to deep water, leaving the edges sandy; i.e., creates a beach.
Soil classification, which is not easily quantified, appeared to reveal more information about the
climatic and hydrologic conditions in wetlands than about land use. A notable exception, however,
occurred at wetland C7 at the CWLSA. In the wet meadow of C7, we found an apparent buried A
horizon and "cumulic," dark-colored A horizons in all 4 wet meadow profiles. Cumulic A horizons are
relatively unusual in wet meadow sites where wave action generally sorts fine-grained soil material and
organic colloids. The other wet meadow profiles we classified in other wetland basins at the CWLSA
had much thinner and sandier A horizons than those at C7. We interpret this difference to relatively
rapid soil deposition on the C7 wet meadow.
For the soils sampled from the wetlands in the 10.4 km2 plots, generally, the eastern wetlands
were non-calcareous Endoaquolls and Argiaquolls, and the more western wetlands were, more
calcareous Typic Calciaquolls, Cumulic (Calc) Endoaquolls and Aerie Calciaquolls. In some cases,
apparent changes in classification from 1992 to 1993 within the same zone (e.g., wet meadow in
wetland 442-295, Appendix 7-1), reflected the drought-to-deluge transition that occurred over much of
the northern prairie from 1992 to 1993. Classification of Mollisols often hinges on the presence or
absence of teachable constituents, especially (CaCO3). In upland soils, where runoff occurs and
leaching rate is relatively low, calcareous soils are probably stable through typical northern prairie
climatic fluctuations. However, where water is focused in the landscape, i.e., in the wetlands, leaching is
apparently capable of removing enough CaCO3 from a soil profile to alter its classification.
7.5.4 Soil Oxidation-Reduction
Soil oxidation-reduction is a function of temperature, microbial activity, and the availability of
elements or compounds that can serve as electron acceptors during metabolic activities. Under aerobic
conditions, oxygen is the primary electron acceptor, but when oxygen is depleted, other compounds
including nitrate, manganese oxides, iron oxides, sulfate, and carbon serve the microbial community as
electron sinks. Oxidation and reduction of iron oxides in water-logged soils produces observable
"redoximorphic" patterns (commonly known as mottles) we use to assess the hydrologic characteristics
of soil subject to saturation. In 1992 we had experienced 6 years of drought and had questions about
whether the redoximorphic features we were seeing at CWLSA were due to contemporary oxidation-
reduction processes, or whether the morphology was relict. The monitored sites became mostly
142
-------
saturated and reduced in the spring of 1993. The sites were under several feet of water by late
summer, 1993 and undoubtedly anaerobic. At that point we know that during reduction iron is
mobilized.
Although important in understanding the link between soil morphology and biogeochemical
processes, the oxidation-reduction data we collected do not appear to be valid condition indicators.
7.6 FUTURE RECOMMENDATIONS
Soil P is the best indicator of wetland condition based on results from our study, specifically, P
found in the 0-15 cm depth. We recommend the continued use of NaHCO3-extractable P (Olsen et al.
1954, Knudsen and Beegle 1988) when testing for biologically available P in northern prairie wetlands.
This test, also commonly known as "Olsen phosphorus" is routinely performed at relatively low cost at
the NDSU Soil Fertility Laboratory.
Organic matter and texture analyses, although not statistically significant variables in this study,
remain potential indicators of severe soil disturbance, and we recommend they be included in future
studies.
EC can show a relationship between fluctuating climate and landscape salinity, but based on
our data we do not recommend it as an indicator of wetland condition. Salinity is a water quality issue
in parts of the PPR, and it may be useful to track long-term precipitation-soil salinity patterns to better
understand how landscape salinity responds to climatic variations. An inexpensive 1:1 soil-water
suspension EC test might be useful for this purpose.
In addition to our doubts about using Cs-137 because of the confounding effect of cultivation, it
is too expensive. Our data indicate a relatively high number of samples from each basin would be
needed to develop accurate results for deposition rates in PPR wetlands. Each gamma sensor can
count only 1 or 2 samples a day, depending on its sensitivity and the amount of Cs-137 in the soil.
Further, since fallout from the 1963-64 maximum is already half of its original activity, we will have less
Cs-137 activity to work with as time goes by. Therefore, we do not recommend Cs-137 for long term
monitoring in the PPR.
143
-------
We recommend dropping the following variables from indicator testing: NO3", pH, calcium
carbonate equivalent test, soil classification, soil oxidation-reduction and Cs-137.
144
-------
Section 8.0
PESTICIDES IN WETLAND SEDIMENTS
AS INDICATORS OF ENVIRONMENTAL STRESS
Diane L. Larson
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
Pesticide use is an established agricultural practice in the northern Great Plains. Grue et al.
(1986) estimated that 80-90% of row crop acreage is treated with herbicides. In a survey of water
quality at streamflow-gaging stations throughout the corn and soybean belt of the U.S., Thurman et al.
(1992) detected atrazine in 98% of postplanting samples; 55% of these detections were above the
maximum contaminant level (MCL) set by the U.S. Environmental Protection Agency.
Pesticides primarily enter wetlands in runoff or as oversprays. For example, spikes in
concentrations of triazines in streams during postplanting runoff events reached an order of magnitude
higher than the MCL, and were correlated with streamflow (Thurman et al. 1992). More ethyl parathion
reached emergent wetland vegetation than the target sunflower plants during aerial spraying trials
(Tomeetal. 1991).
Until the development of enzyme-linked immunosorbent assay (ELISA) techniques for detecting
and quantifying pesticides in water and sediment, broad-scale screening for pesticides could be
prohibitively expensive. In this study, I examine the potential use of ELISA-determined pesticide levels
in wetland sediments as a measure of wetland condition.
8.1 OBJECTIVES
To assess the utility of pesticide levels in wetland sediments for discriminating between good
and poor quality wetlands.
145
-------
8.2 METHODS
Hal Kantrud collected sediment samples during the course of his vegetation sampling in July
1992 and 1993. He collected three 5-g samples from the innermost portion of each wetland basin, one
each from quadrats 1, 3 and 5 (see section 6.2 for sampling framework). Samples were placed
individually in Zip-Loc bags and kept in ice-filled coolers while the researchers were in the field. Each
Zip-Loc bag was labeled with the wetland identification number and date. Upon returning to Northern
Prairie Science Center, the crew placed the samples in a refrigerator, where the sediments were kept
chilled, but not frozen. After all samples had been collected, they were packed in plastic coolers and
sent by overnight service to USGS in Bismarck for analysis.
In 1992 we analyzed all samples, regardless of surrounding cropland composition, for atrazine.
In 1993, Hal Kantrud recorded the composition of cropland around each wetland basin (see Dwire
1994 for data sheet), and we analyzed for 2,4-D in wetlands near small grain fields and for cyanazine
in wetlands near corn fields. Wayne Berkas, Water Quality Specialist with U.S.G.S., conducted the
analyses using RaPID Assay® Kits from Ohmicron (see Dwire 1994 for lab procedures). Ten percent
of the samples were replicated.
8.3 RESULTS
8.3.1 Atrazine
Atrazine is a herbicide widely used in corn production. No other crops within the EMAP study
area are sprayed with atrazine. The only strata in which corn is extensively grown are Low-North and
Low-South, so this analysis is limited to these strata. Atrazine was found more often and in higher
concentrations in wetlands classified as poor condition than in those classified as good condition in both
Low-North and Low-South strata (Table 8-1).
8.3.2 2,4-D
The herbicide 2,4-D is primarily used in small grain production, and is the most common
herbicide in use in North Dakota (Grue et al. 1986). Nonetheless, we were relatively unsuccessful in
detecting 2,4-D in wetland sediments. Of 32 EMAP wetlands tested for 2,4-D, only four had values
146
-------
above the 15 ppm detection limit in even one of the three samples taken from each wetland; none had
values higher than 17 ppm. Three of these wetlands were within small grain production areas; the
fourth was near a Waterfowl Production Area that had been sprayed recently for thistle.
8.3.3 Cyanazine
In 1993, we intended to test for cyanazine, a herbicide most often used in cornfields, but
because Low-North and Low-South strata were dropped we no longer had an adequate sample of
wetlands in areas of corn production. We did test for this herbicide in conjunction with research
described in Section 9.2. Briefly, we were unsuccessful in detecting cyanazine in any wetland sediment.
Table 8-1. Atrazine concentration in wetland sediments determined by ELISA. The lower detection
limit was 15 ppb.
Plot Site Stratum Health Atrazine
38
38
38
54
54
54
59
59
60
60
60
60
60
241
241
246
246
246
246
246
246
249
249
249
396
396
396
62
63
62
39
39
39
42
111
58
58
128
128
128
3
48
34
34
37
37
53
53
50
50
86
107
107
130
LN
LN
LN
LN
LN
LN
LN
LN
LN
LN
LN
LN
LN
LS
LS
LS
LS
LS
LS
LS
LS
LS
LS
LS
LS
LS
LS
L
L
L
L
L
L
L
H
H
H
H
H
H
L
L
L
L
L
L
L
L
H
H
H
H
H
H
26
21
33
0
0
0
0
0
0
0
0
0
0
0
122
0
15
28
0
0
0
0
0
0
0
0
16
8.4 EVALUATION AND RECOMMENDATIONS
In evaluating the potential use of ELISA-determined pesticide levels in wetland sediments as a
measure of wetland condition, extent of use, persistence in soils, and availability of test kits must be
147
-------
taken into account. 2,4-D is the most extensively used of all herbicides in the northern Great Plains, but
it degrades very rapidly. Field tests indicate 90% dissipation at the soil surface in as little as 40 days
(Nash 1988) Sampling would have to be coordinated closely with the small grain planting season to
optimize detection. For this study, samples were collected in July; early June may be a more
appropriate time.
Atrazine is relatively persistent in the environment, with 90% dissipation at the soil surface
taking as much as 140 days (Nash 1988), and thus allows for less precision in sampling date. The
herbicide is ubiquitous in areas of corn cultivation (Thurman et al. 1992), and concentrations in
sediments discriminate well between good and poor condition wetlands in these areas (Table 8-1). In
only one case was atrazine detected in a wetland classified as good condition, suggesting little chance
of falsely assigning poor condition to a good condition wetland.
148
-------
Section 9.0
DEVELOPMENT OF NEW SAMPLING METHODS
AND SAMPLING TECHNIQUES
9.1 DEVELOPMENT AND EVALUATION OF AN INVERTEBRATE SAMPLING
DEVICE AND A WATER-LEVEL RECORDER FOR EMAP
Ned H. Euliss, Jr. and David M. Mushet
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
9.1.1 Introduction
Aquatic invertebrates are potentially valuable indicators for EMAP because they are highly
sensitive to environmental change, especially those induced by agricultural practices. However,
invertebrates are so highly variable in space and time that single measurements may fail to detect
important changes. At the onset of this pilot study, there were no techniques available that would permit
collection of invertebrates in a manner compatible with EMAP objectives. One objective of this study
was to develop and evaluate an invertebrate sampling device that collected time-integrated information
on macroinvertebrates. We used sediment traps described in the scientific literature to sample
recalcitrant remains of invertebrates over discrete time periods. We correlated the abundance and
biomass of invertebrates, determined from remains captured in the sediment traps, with population
estimates obtained using more labor intensive sampling methods (monthly sweep-net sampling) to
evaluate the possibility of using sediment traps to sample invertebrate populations for EMAP.
Removal of grasses and other native vegetation from wetland watersheds alters surface runoff
dynamics and hence exacerbates impacts associated with sedimentation and agricultural chemicals
adsorbed on soil particles. Nonvegetated watersheds have less capacity to mitigate excessive surface
runoff, resulting in water levels in wetlands that are more variable than in wetlands in landscapes
dominated by grasses and forbs. Thus, fluctuations in water levels may prove to be a valuable indicator
of wetland condition. However, the dynamic hydrology of prairie wetlands is difficult to measure and
currently requires the use of continuous-recording, water-level monitors. The high costs of these
devices usually precludes their use except on a very limited basis. An additional objective of this study,
149
-------
was to develop and evaluate an inexpensive device for recording maximum and minimum water levels
in wetlands.
9.1.2 Objectives
1. Develop and evaluate a quantitative device that samples aquatic macroinvertebrates
indirectly by capturing their recalcitrant remains.
2. Determine if standard sediment traps yield sediment dry weights useful as indicators of
wetland condition.
3. Develop and evaluate an inexpensive water-level recorder for determining wetland
condition.
9.1.3 Methods
9.1.3.1 Objective 1
From April 1992 to September 1993, we evaluated five prototype devices, previously described
to collect sediments (Gardner 1980), as collectors of recalcitrant remains of aquatic invertebrates. This
study was not conducted in the EMAP pilot study wetlands used by other EMAP investigators due to
the intensive sampling required to determine standing crops of selected aquatic macroinvertebrates.
Instead all sampling was conducted in 18 randomly chosen semipermanent wetlands located in
Stutsman county, ND (Table 9-1). Wetlands were selected in both highly impacted landscapes (poor
condition) and marginally impacted landscapes (good condition). To avoid landowner reluctance to
provide access, most wetlands selected were on Waterfowl Production Areas (WPA's) owned by the
U.S. Fish and Wildlife Service. Most WPA's are not farmed, however, poor condition wetlands existed
along property boundaries where a portion of the watershed was outside of the WPA's boundaries.
We designed 4 prototype sampling devices that were modifications of devices previously
described for collecting sediments from lentic waters (Bloesch and Burns 1980; Garner 1980; Blomqvist
and Hakanson 1981) (Fig. 9-1). Each device consisted of a 51 cm long piece of 2" (5.1 cm) inside
diameter I.D. PVC pipe (collection tube) that was capped at one end with a standard 2" PVC cap. Thus,
150
-------
.each device had an aspect ratio of 10 to facilitate the least biased estimate of downward sediment flux
(Hargrave and Burns 1979; Lau 1979; Bloesch and Burns 1980; Garner 1980; Blomqvist and Hakanson
1981).
Table 9-1. Legal descriptions of tracts of land containing wetlands used to evaluate quantitative
devices that sample recalcitrant remains of selected aquatic macroinvertebrates and
sediment deposits. All wetlands are located within Stutsman County, North Dakota.
Wetland
Number
13e
16
20
281
2BII
28III
39a
39GI
39GII
48
54
98a
106
122
154
421
462a
462b
Legal Description
NE
NE
NE
NE
NE
NW
SW
NE
SE
SE
SE
NW
SW
NW
SE
NW
NE
SE
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
1/4,
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
33,
9,
13,
4,
4,
3,
34,
32,
32,
12,
35,
5,
23,
35,
5,
2,
34,
35,
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
Township
14 2N,
14 IN,
139N,
14 IN,
14 IN,
14 IN,
14 2N,
142N,
14 2N,
139N,
137N,
139N,
14 ON,
14 IN,
143N,
13 9N,
139N,
139N,
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
Range
68W
68W
67W
66W
66W
66W
66W
66W
66W
66W
67W
67W
68W
67W
63W
68W
65W
65W
The four sediment trap types differed from each other only in the size of the opening into the
trap or in the placement of the trap relative to the sediment/water interface. The first device (straight-
tube trap) consisted of simply the PVC collection tube capped at one end and with no modifications to
the open end. The second device (funnel-top trap) was similar to the straight-tube trap except a 2" (5.1
cm) X 4" (11.4 cm I.D.) PVC bell adapter was glued to the open end of the collection tube. The third
device (bottle-top trap) was similar to the funnel-top trap except instead of the bell adapter a 2" (5.1
cm) X 3/4 inch (1.9 cm) pipe, PVC reducer was glued to the open end and a 3/4 inch pipe X 3/4 inch
tubing (1.5 cm I.D.) adapter was screwed into the reducer's opening. Thus, the opening into the
collection tube of the bottle-top trap was reduced to 1.5 cm. The straight, funnel-top and bottle-top traps
were all installed vertically in the wetland sediments so the top (open end) of the trap extended 7.4 cm
above the water/sediment interface (Fig. 9-1). The fourth device (flush trap) was a replicate of the
straight-tube trap but installed so that the top of the trap was flush with the water/sediment interface.
151
-------
The flush traps collected too much material and quickly filled with sediment and organic debris.
Because much of this material was unconsolidated sediments, flush samples were dropped from further
testing.
We installed the sediment traps at sampling stations established within each of the 18 replicate
wetland basins in May and removed them in September of each year (1992 and 1993). At each station,
we installed one replicate of each trap spaced 60 cm apart in a square pattern (Fig. 9-2). In 1992, we
established four sampling stations located at random locations along transects that radiated from the
center of the wetland along random compass bearings.
In 1993, we established five sampling stations in each wetland basin that were also located on
random transects. However in 1993, the sampling stations were located at specific locations,
corresponding to precise elevations (+1.6 mm per 30 m) that facilitated sampling when the water depth
at the wetland basin center was > 10 cm (Fig. 5-1) using a Spectra-Physics Model 650 Laserplane.
This placed the traps close to the center of the wetland basin to facilitate sampling during periods of
low water. Each year we also installed a feldspar clay marker (Cahoon and Turner 1989) in the center
of the square formed by the four traps to facilitate coring at later dates to sample invertebrate remains
and sediments that accumulated over specific time frames. We experienced problems with our feldspar
marking that precluded their use in our evaluation. Specifically, feldspar clay moved downward in
sediments mostly composed of organic debris and hence was an unreliable measure of sediment
accretion in the Prairie Pothole Region.
In September of each year, we removed the traps from the wetland basins and stored them in
freezers until processing. We processed samples by removing a sample from the collection tube while it
was still frozen, sieving the thawed sample residue on a 0.5 mm sieve, and separating the invertebrate
recalcitrant remains (i.e. cladocera ephippia, ostracod shells, conchostracan shells, and gastropod
shells) from sediment debris using light tables and forceps. All sample residues were retained during
the sieving process and soil and other debris > 0.5 mm remaining after removal of invertebrates was
returned to the sediment sample previously screened for later determination of sediment dry weights.
Beginning when sediment traps were placed in study wetland basins and continuing monthly
throughout the open-water, ice-free portion of the year, we collected samples of aquatic
macroinvertebrates from each study wetland basin at each sampling station using 2 foot net sweeps
(Swanson et al. 1974). Samples were preserved in 80% ethanol and transported to the laboratory for
processing. Processing sweep-net samples consisted of straining the sample through a 1 mm mesh
152
-------
7X1 cm
B
G -WetlandSediments
Figure 9-1. Sampling devices tested for EMAP pilot study in 18 wetlands in Stutsman County, ND (A =
Straight-tube trap, B = Flush trap, C = Bottle-top trap, D= Funnel-top trap).
screen to remove excess ethanol, removing aquatic invertebrates from the sample using a light table
and forceps, sorting invertebrates into taxonomic groupings, and enumerating and weighing them to the
nearest milligram on an analytical balance after drying to a constant weight in a drying oven at
55-60 °C; only those taxa that had recalcitrant body parts were considered.
Statistical Methods (Invertebrate analysis). We determined if correlations existed between
the abundance and biomass of invertebrate remains captured in the sediment traps and the abundance
and biomass of invertebrates actually present in the wetlands (determined from sweep-net samples)
using SAS (SAS Institute, Inc. 1989). The purpose of this analysis was to identify the sediment trap that
was most closely correlated with the more labor intensive and standard sweep net samples. In addition,
we also used linear regressions (SAS Institute Inc. 1989) to determine if macroinvertebrate abundance
or biomass, as estimated by the various sediment trap types, could be used to predict the percentage
153
-------
of grassland remaining within wetland drainage basins (the basis of the condition definition used by
other researchers in this pilot study). In 1992, sediment traps were placed randomly along transects
rather than at precise elevations. Extremely dry conditions in 1992 resulted in our study wetland basins
going dry, and as the pool levels dropped, individual sediment traps stopped sampling at various times.
This caused excessive variability among our samples and interfered with our analysis. Therefore, we
used only 1993 data in the statistical analysis of invertebrate abundance and biomass, and sediment
dry weights.
9.1.3.2 Objective 2
Prototype sampling devices also were evaluated to determine the optimal design to monitor
sedimentation for EMAP. We centrifuged all residues from collected samples after invertebrate remains
were removed for Objective 1 at 5,000 rpm for 10 minutes to separate the sediments from excess
water. We then dried the sediments collected by each trap in an oven at 100 C° until a constant weight
was reached and weighed them to the nearest 0.01 g.
Statistical methods. We used linear regression (SAS Institute Inc. 1989) to determine if
sediment dry weight could be used to estimate the percentage of grasslands within each wetlands
watershed and thus serve as indicators of wetland condition. We used the dry weights only from 1993
and estimated regression lines for each trap type.
9.1.3.3 Objectives
We designed a water-level recorder that would provide the maximum and minimum water level
of a wetland basin over discrete time periods. The device consisted of a commercially available,
copper-coated steel welding rod that guided a large float up and down as water levels fluctuated (Fig.
9-3). Two magnetic slides, one above and one below the float were pushed by the float to positions on
the rod that corresponded to the maximum and minimum water levels, respectively (Fig. 9-4). The
distance between the slides was the distance the water level fluctuated during the time period between
installation in the wetland and reading of the water levels. After recording, the device was easily reset
by sliding the magnetic indicators to positions directly above and below the current level of the float.
We installed water-level recorders in two semi-permanent wetland basins (P7 and P8) at the
Cottonwood Lake Study Area (Swanson 1987) in May, 1992. The wetland basins were also equipped
154
-------
with Teiog model WLS-2109 water-level monitoring systems that provided continuous recordings of the
water levels in the 2 wetlands throughout the study period. The Telog recorders were housed inside a
steel pipe that was sunk into the wetland sediment approximately 1 m below the water/sediment
interface. Each unit was recalibrated to read "0" in that position, and the units were turned on for
continuous readings. Although the units are warranted to withstand -40 P, the manufacturer felt burying
the units beneath the wetland soil surface and protecting the transducers from silt deposits by housing
in a pipe, would avoid unanticipated complications and extend their effective life. At the end of each
year, we compared the water-level fluctuation (maximum depth - minimum depth) recorded by our
prototype devices to that determined from the data collected from the water-level monitoring systems.
9.1.4 Results
9.1.4.1 Objective 1
Our correlation analysis of macroinvertebrate abundance with bottle-top, funnel-top, and
straight-tube sediment traps was performed only on Cladocera ephippia, Ostracods, Conchostracans,
and 3 Gastropod taxa (Planorbids, Physids, and Lymnaeids). While the correlations were clearly unique
for each taxon, all appeared to be adequately sampled by either the straight-tube or the funnel-top
sediment traps (Table 9-2). Further, straight-tube sediment traps yielded significant correlations for
Conchostracans, Planorbid snails, and Lymnaeid snails that were higher than correlations with other
trap types. Cladocera ephippia were most correlated (r=0.593) with funnel-trap samples but the
correlation with the straight-tube trap also was significant and had a nearly identical correlation
(r=0.584) (Table 9-2). Similarly, Physid snails were most correlated with the funnel-trap samples
(r=0.910) but the straight-tube trap was also yielded a significant correlation (r=0.785). Only for
Ostracods was the funnel-top trap, the clear choice for sampling an invertebrate taxon; it yielded the
only significant correlation (r=0.601) of any trap types considered. Bottle-top sediment traps did not
estimate any invertebrate taxon better than other trap types.
Our correlation analysis between macroinvertebrate biomass and the three sediment trap
designs yielded results that were generally consistent with our findings for invertebrate abundance
(Table 9-3). Straight-tube samples were highly correlated with Cladocera ephippia (r=0.798),
Conchostracans (r=0.913), Planorbid snails (r=0.671), and Lymnaeid snails (r=0.452). Although not
significant, the highest correlation (r=0.452) for Physid snails also was with the straight-tube sediment
trap. As was the case with the abundance analysis, the correlation between Ostracods and the
155
-------
Wetland Edge
1992
Random
Distance
1993
Precise
Elevation
wetland Center
Figure 9-2. Configuration of sampling stations located on random transects (A=Straight-tube trap, B=Rush
trap, C=Bottle-top trap, D=Funnel-top trap, and F=Feldspar clay).
156
-------
3" PVC Cap
3" X 31
PVC Pipe
Magnet
Maximum
Indicator
Float
1/8" X 31
Copper Clad
welding Rod
Minimum
Indicator
3" PVC cap
2.54 cm
T
6.5 cm
Figure 9-3. Prototype water level recorded designed for EMAP pilot study to measure water depth
fluctuations in wetland basins.
157
-------
= Float I » Indicator
Water
Figure 9-4. Diagram of prototype water level recorder showing how changes in water levels move the
float and thus the indicators providing a measurement of water level fluctuation.
straight-tube trap was low and nonsignificant. However, and as was the case with the abundance
analysis, the best correlation for Ostracods was with the funnel-top trap, although the correlation was
nonsignificant in the biomass analysis. In contrast with the abundance analysis, the best correlation for
Lymnaeid snails was with the bottle-top trap, although the correlation with the straight-tube trap was
significant as well.
In our analysis to determine if invertebrate abundance or biomass (as determined from remains
captured in sediment traps) could be used to estimate the proportion of the watershed remaining in
grassland, we failed to reject the null hypothesis in all cases (Table 9-4).
158
-------
Table 9-2. Correlations with probability values (P-value) in parentheses between abundance of
invertebrate remains captured in 3 types of sediment traps (bottle-top, funnel-top, and
straight-tube) and invertebrate abundance of wetlands determined from monthly sweep-net
samples with most influential observations removed,1993.
Correlations (n = 18)
Taxon
Cladocera
Ostracoda
Conchostraca
Pianorbidae
Physidae
Lymnaeidae
Bottle-Top
r
0.
0.
0.
-0.
0.
0.
365
500
633
132
088
520
P
(0
(0
(0
(0
(0
(0
.1499)
.0408)
.0064)
.6015)
.7468)
.0324)
Funnel-Top
0
0
0
0
0
0
r
.593
.601
.448
.335
.910
.843
P
(0
(0
(0
(0
(0
(0
.0197)
.0137)
.9716)
.2217)
.0001)
.0001)
Straight
r
0
-0
0
0
0
0
.584
.058
.701
.494
.785
.918
P
(0.0175)
(0.8185)
(0.0017)
(0.0372)
(0.0001)
(0.0001)
Table 9-3. Correlations with probability values (P-value) in parentheses between biomass of
invertebrate remains captured in 3 types of sediment traps (bottle-top, funnel-top, and
straight-tube) and invertebrate biomass of wetlands determined from monthly sweep-net
samples with most influential observations removed, 1993.
Correlation
Taxon
Cladocera
Ostracoda
Conchostraca
Pianorbidae
Physidae
Lymnaeidae
Bottle-Top
r
0
0
0
0
0
0
.452
.041
.738
.091
.097
.777
P
(0
(0
(0
(0
(0
(0
.0688)
.8805)
.0007)
.0727)
.7211)
.0002)
(n = 18)
Funnel-Top
0
0
0
0
0
0
r
.296
.185
.628
.811
.062
.498
P
(0
(0
(0
(0
(0
(0
.2859)
.4771)
.0069)
.0001)
.8177)
.0354)
Straight
r
0
0
0
0
0
0
.798
.053
.913
.671
.452
.470
P
(0.0002)
(0.8387)
(0.0001)
(0.0032)
(0.0683)
(0.0493)
Table 9-4. Results of linear regressions to determine if macro!nvertebrate abundance or biomass, as
estimated by the various sediment trap types, could be used to predict the percentage of
grassland remaining within each wetland's drainage basin.
Trap Type
Bottle-top
Funnel-top
Straight-tube
Predictor
Abundance
Biomass
Abundance
Biomass
Abundance
Biomass
T-statistic
1.035
0.852
0.316
-0.376
0.362
0.577
P-value
0.3162
0.4070
0.7565
0.7116
0.7223
0.5718
159
-------
9.1.4.2 Objective 2
We were unable to find a significant relationship between the proportion of a wetland's
watershed in grassland and the dry weights of sediments collected in the straight-tube traps (T = 0.903,
P =0.3797), the funnel-top traps (T = 0.685, P = 0.5030), or the bottle-top traps (T = 1.374, P =
0.1885).
9.1.4.3 Objectives
The two water-level monitoring systems in the Cottonwood Lake wetlands functioned properly
over the 2 year time period of this study. In 1992, the water level of wetland P8 peaked on July 2 at its
maximum depth for the year of 25.9 cm, and then dropped steadily until the wetland went dry on
August 18 (Fig. 9-5). Wetland P7 followed the same trend, reaching a maximum depth of 28.2 cm on
June 6 and steadily falling to 0.0 cm on July 23 (Fig. 9-6).
In 1993, because of heavy rainfall, the trends were reversed with both wetlands steadily gaining
water throughout the summer (Figs. 9-5 and 9-6). Wetland P8 had a minimum water depth of 63.6 cm
on May 6 and the depth increased to a maximum of 136.7 cm on July 25. Wetland P7 reached a
minimum depth of 39.6 cm on April 5 and increased to a maximum of 131.3 on August 31.
The water-level recorders we designed accurately recorded (+ 1.4 cm) the maximum and
minimum water levels of the two Cottonwood Lake wetlands as determined by the Telog water-level
monitoring systems both years except for 1993 when the maximum levels of the wetlands exceeded
the capacity of our prototype recorders (Table 9-5 ).
9.1.5 Evaluation
9.1.5.1 Objective 1
The correlation analysis of macroinvertebrate abundance and biomass with the three trap types
generally identified the straight-tube traps as providing the best correlations with the much more labor
intensive and costly method of collecting monthly, sweep-net samples. The only caveat is that the other
160
-------
Wetland P8
Legend
1882
1993
11 May 1 June 1 July 1 August 1 Sept. 1
Figure 9-5. Water levels recorded with Telog water level monitor of wetland P8 at the Cottonwood Lake
Study Area, Stutsman County, ND, April to September, 1992 and 1993.
traps may provide better representation of specific taxa (e.g., funnel-top traps for Planorbid snail
biomass). Bottle-top traps were useful only for Lymnaeid snail biomass although both funnel-top and
straight-tube traps yielded significant, albeit lower correlations. In general, we recommend that straight-
tube traps be utilized for all taxa except ostracods, unless specific taxa can be identified as indicators of
wetland condition.
Our regression analysis did not differentiate between wetland basins in good or poor condition
using either macroinvertebrate abundance or biomass. While it is highly likely that agricultural practices
161
-------
Wetland P7
140
ril 1 May1 Junel Julyl August 1 Sept. 1
Figure 9-6. Water levels recorded with Telog water level monitor of wetland P7 at the Cottonwood Lake
Study Area, Stutsman County, ND, April to September, 1992 and 1993.
do impact the invertebrate community, it was not observed in the taxa collected by our sediment traps.
It should be noted that our sampling devices collected a very small proportion of the invertebrates
present in wetlands because only those taxa that have recalcitrant body parts (i.e., ostracods,
conchostracans, and gastropoda) or have easily identified resting eggs (i.e., Cladocera ephippia) were
represented in our samples. Studies have shown that certain taxa (e.g., amphipods, chironomid
midges) are highly susceptible to agricultural practices, especially chemical application (Grue et al.
1989; Tome et al. 1990). However, those taxa were not collected by our sampling devices and hence
were not evaluated as potential indicators of wetland condition in this study.
162
-------
Table 9-5. Maximum and minimum water levels (cm) of wetlands P8 and P7 at the Cottonwood Lake
Study Area, Stutsman County, North Dakota, as recorded by prototype water-level
recorders developed for the EMAP pilot study.
1992
Wetland
P8
P7
Maximum
25.4
27.9
Minimum
0.0
0.0
1993
Maximum Minimum
* 65.0
* 39.4
* = Maximum water depth > capacity of gauge.
Recommendations for future work. Aquatic invertebrates clearly have potential as indicators
of wetland condition in the Prairie Pothole Region. However, the sediment traps used to quantify
invertebrates in this study did not appear to sample taxa that would serve as useful indicators of
wetland condition. We recommend that future EMAP studies continue to explore invertebrates as
potential indicators of wetland condition. However, because of the limitations identified by our study, we
recommend that sediment traps be dropped as a quantitative tool, although trap designs were identified
that provided reliable measurements of both invertebrate abundance and biomass. Further, we
recommend that within-basin sampling be dropped from consideration because of the high natural
variability that characterizes invertebrate fauna, both on spatial and temporal scales and because of the
high variability we observed between wetlands. In the sample size analysis we conducted in Section 5
of this report, it was shown that we collected too few samples from an insufficient number of wetlands.
Although we did not perform a similar analysis in this section, there was extreme variability in these
samples as well, and it is likely that we again collected too few samples from an insufficient number of
wetlands. While it is possible to increase the intensity of the sampling effort, it is not practical in terms
of material and labor costs. We feel that the most viable approach would be to focus on landscape-level
measures of invertebrate richness, abundance, or biomass. Conceivably, such an approach is possible
by using either light traps or sticky traps (Belton and Kempster 1963, Harding et al. 1966, Belton and
Pucat 1967, Mason and Sublette 1971, Davidson et al. 1973, and Borror et all. 1981) to collect samples
uniformly from the hexagon over prolonged periods of time.
9.1.5.2 Objective 2
Our analysis of sediment dry weights also failed to identify an indicator of wetland condition.
Although sedimentation is clearly a major impact on wetlands in the Prairie Pothole Region (Gleason
163
-------
and Euliss, unpublished data), we were unable to find a significant relationship between the proportion
of grassland within wetland watersheds and the dry weights of sediment in any of the sediment trap
designs we evaluated. One caveat of this analysis, was that traps were placed in the deepest portion of
wetland basins to optimize sampling for aquatic invertebrates rather than for sediments. Wetlands tend
to silt in from the sides; hence, trap placement was likely not optimal to accurately measure silt loads
washing into wetland basins. As found in Section 5 of this report, there was substantial variability, both
within and among wetlands in our samples and hence we may have collected an inadequate number of
samples. However, we feel that the most important factor affecting our results was where the sediment
traps were spatially located within wetlands.
Recommendations for future work. We feel that it is intuitive that wetland landscapes heavily
impacted by agriculture experience increased rates of sedimentation and that our negative results were
strongly influenced by trap placement. There was no official study of sedimentation in this pilot EMAP
study, and the limited work that was done here was facilitated through a collaborative effort with John
Freeland (Section 7 of this report). Because the budget for the invertebrate work in the pilot study
included the sediment traps, they were spatially placed in areas that optimized the sampling for
invertebrates rather than for sediments. In future work, we recommend that sediment collection devices
be situated near the periphery of wetlands where it is much more likely that siltation events can be
recorded. A separate and unrelated study of wetland siltation on the Woodworth WPA, Stutsman
County, ND, has clearly documented elevated sedimentation rates in wetlands with tilled watersheds
relative to wetlands that have grassland watersheds (Gleason and Euliss, unpublished data). Further,
we recommend that surface flow traps (Gleason and Euliss, unpublished data) be utilized in any further
indicator development and evaluation effort to avoid problems associated with within-basin phenomenon
such as resuspension of sediment due to cattle grazing, wind action, and other events not directly
related to land-use within the watershed.
9.1.5.3 Objective 3
As indicated in our original study proposal, there were only two continuous water-level
monitoring systems purchased, and the data were too few to statistically analyze. However, the devices
have been extensively tested by the manufacturer and are generally accepted quantitative tools. In
1993, we received record rainfall that flooded area wetlands to excessive depths and the capacity of
our prototype devices was exceeded and the seasonal maximum was not recorded. Our continuous
monitoring system recorded maximum pool levels of 137.7 cm and 131.3 cm for wetlands P8 and P7,
164
-------
respectively. The prototype water-level recorders failed because the devices were only 91 cm in length
and were over-flooded when the water levels rose to unanticipated levels.
Recommendations for future work. While it would be highly desirable to equip all EMAP
wetlands with continuous water-level monitoring systems, their high cost will preclude their use except
on a very limited scale. The units evaluated in this study performed flawlessly and in accordance with
the manufacturers specifications, despite being exposed to extremely frigid winter temperatures
followed by spring thawing. The water-level recorders designed in this study and tested in Section 5 of
this report clearly demonstrate that water-level fluctuation, when divided by the area of the wetland's
watershed, is a useful indicator of wetland condition. Thus, we recommend that any future indicator
studies in the PPR include these devices instead of the more expensive water-level monitoring
systems. However, we recommend that future indicator research in the PPR construct water-level
recorders that are longer than the ones used in this study to avoid missing maximum water levels in the
event of excessive flooding and that less corrosive materials (see Section 5 for discussion) be used in
their construction.
9.2 HORMONAL RESPONSE TO ENVIRONMENTAL STRESS: TECHNIQUE
DEVELOPMENT
Diane L. Larson
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
9.2.1 Introduction
Harlow et al. (1987) have defined an animal in "stress" as one that is "required to make
abnormal or extreme adjustments in its physiology or behavior to cope with adverse aspects of its
environment." Physiologically, stress in vertebrates is accompanied by an increase in plasma
corticosteroid levels (Harlow et al. 1987, Kirkpatrick et al. 1979, Licht et al. 1983, McDonald et al. 1988,
McDonald and Taitt 1982, Moore and Deviche 1987, Moore and Miller 1984, Orchinik et al. 1988, Seal
and Hoskinson 1978, Whatley et al. 1977, Wingfield et al. 1982). Often such increases accompany a
decline in immune system response which may make stressed populations more susceptible to disease
(Seller and Christian 1982) and parasitism. Heart rate in domestic sheep is positively correlated with
corticosterone levels (Harlow et al. 1987), suggesting a generally higher cost of metabolism under
stress. High levels of corticosteroids have also been associated with decreased or abolished
165
-------
reproductive behavior in amphibians (Dupont et al. 1979, Moore 1983, Moore and Deviche 1987) and
fish (Campbell et al. 1994).
Parsons (1990) has pointed out that the affect of individual environmental stressors cannot be
considered in isolation. Stressors such as environmental contaminants are often difficult and expensive
to measure, and their potential synergisms are largely unknown; measures of the level of individual
contaminants may underestimate their effect on the organism. Because corticosteroid release is a
common response to a range of stressors across a wide variety of taxa, measures of corticosteroid
levels may provide an index to the amount of stress a population is experiencing. Because the
response is non-specific, no assumptions are necessary regarding the cause of the stress.
Corticosterone levels may respond to environmental stress in two ways. Baseline levels may
become persistently elevated. This response has been observed in the reptiles Lacerta vivipara
(Dauphin-Villemant and Xavier 1987) and Urosaurus ornatus (Moore et al. 1991). However, baseline
levels are difficult to measure with certainty because the onset of the stress response is usually quite
rapid. In addition, baseline levels may vary seasonally, and with the age and sex of the animal.
Recent work on birds (J.C. Wingfield, University of Washington, pers. comm.) and amphibians
(F.L. Moore, Arizona State University, pers. comm.) has indicated that chronic (e.g., environmental)
stress may also affect the rate at which corticosterone levels respond to acute stress. Hontela et al.
(1992) found that the cortiso! stress response was abolished in fish taken from environments polluted
with RGB's. A more appropriate potential indicator of chronic stress thus may be the change in
corticosterone levels in response to acute capture stress.
Because the area of interest in this study is wetland habitat in the prairie pothole region, larval
tiger salamanders (Ambystoma tigrinum) provide an appropriate organism for study. Larvae breathe
primarily through gills and cannot leave their natal wetland until after metamorphosis. In North Dakota,
eggs are laid on submerged portions of emergent vegetation in early spring; larvae begin to hatch as
water temperatures exceed 10 °C. Metamorphosis may occur from August into September, depending
on ambient conditions.
166
-------
9.2.2 Objectives
Objectives of the study were (1) to develop a non-destructive technique for assessing hormonal
response to acute stress under field conditions (addressed in 1992); and (2) to assess the relation
between acute and chronic stress responses (addressed in 1993), under the hypothesis that larvae
inhabiting wetlands in poor condition would experience chronic stress and thus show a diminished
response to acute stress.
9.2.3 Methods
Site selection. In 1992, we were unable to find any salamander larvae in the wetlands included
in the EMAP sample, most of which were dry. Because the primary goal was technique development,
the decision was made to test capture and sampling methods at wetlands in which populations of tiger
salamanders were known to exist. Such a wetland was found in Barnes County, ND, at which most of
the sampling was done. We also sampled larvae from known populations in Kidder County, ND, and in
Deuel and Brookings Counties in SD to examine geographic variability in response.
Based on results from 1992, we restricted our wetland selection to semipermanently-flooded
basins in 1993. To avoid the logistical problems associated with obtaining landowner permission on
another large set of wetlands, we chose as our potential sample pool all semipermanent wetlands in
WPA's in Stutsman and Kidder Counties. During July, 1993, we visited all wetlands within this pool.
Traps were set for one night in each accessible semipermanent wetland with any open water to
ascertain presence of salamander larvae. Of 25 wetlands in which larvae were found, 19, drawn at
random, were re-visited in August. (See Appendix 9-1 for maps showing locations of wetlands sampled
in 1993.)
Capture and handling. We captured larvae in unbaited funnel traps left in wetlands overnight.
Over a one-month period during which larvae were large enough to provide blood samples but not yet
beginning metamorphosis, we obtained blood samples from 4 populations in 1992 and from 19
populations in 1993. Blood samples were taken immediately from half the captured animals removed
from traps; the remaining animals were sampled after 20- or 45-min (in 1992) or 30-min (in 1993)
confinement in 500-ml bottles (acute stress). Larvae were marked and released at the point of capture;
marked larvae were not resampled. See Dwire (1994) for details of sampling procedures.
167
-------
Wetland chemistry and surrounding land use. In 1993, we recorded water temperature, pH,
alkalinity, dissolved oxygen, Carbon dioxide, chloride, hardness, and ammonia in the 19 wetlands in
which we sampled salamander larvae, and in 26 additional wetlands that did not contain larvae in 1993.
We used a LaMotte Test Kit and processed samples in the field. Three 50-g sediment samples also
were collected from each wetland, near where the traps had been deployed. These samples were sent
to USGS in Bismarck to be analyzed for 2,4-D or cyanazine (see Section 8.1 for pesticide analysis
techniques).
Laboratory exposures. We exposed tiger salamander larvae, captured in ND and shipped by
overnight service to the EPA Environmental Research Laboratory in Corvallis, OR, to
Guthion-contaminated or clean aquaria for 10- or 20-day periods. After exposures, half the animals
were subjected to acute stress by confinement in a 500 ml bottle before blood was sampled by
decapitation. The other half of the animals were sampled immediately upon removal from the aquaria.
Results of these experiments were used to compare field and laboratory corticosterone levels, and to
assess acute and chronic stress response under controlled conditions. The brains of the animals were
analyzed for brain cholinesterase, to evaluate Guthion as a chronic stressor. Details of the study plan
can be found in Appendix 9-2.
9.2.4 Results
Technique development (1992). Corticosterone levels were significantly lower in larvae from
which blood was drawn immediately than in those subjected to acute stress (Fig. 9-7). Means are
statistically different: F = 4.877; df = 2, 150; P = 0.009 (one-way ANOVA). Thus, the capture technique
did not obscure the acute stress response. Furthermore, we found a consistent pattern of acute stress
responses among populations from the different wetlands sampled in 1992.
Field test. Of the 19 WPA wetlands we sampled in 1993, 9 were on the Missouri Coteau (refer
to map, Fig. 1-1); none of these had any cropland adjacent to them. Coteau wetlands were therefore
designated "good condition". The remaining 10 wetlands were in the drift plain, east of the coteau, and
all had at least some adjacent cropland (Table 9-6). Wetlands located in the drift plain (Fig. 1-1) were
designated "poor condition". Salamanders on the Missouri Coteau were of the subspecies
melanostictum; salamanders on the drift plain are likely of the subspecies tigrinum. We could not
distinguish occupied from unoccupied wetlands based on water chemistry, presence of 2,4-D or
cyanazine, or land use.
168
-------
Response to acute stress was not consistent between larvae in good- and poor-condition
wetlands. For populations in the agriculturally-impacted drift plain, magnitude of the acute stress
response tended to be inversely related to the percentage of adjacent land in crops (R2=0.30, F=4.23,
p=0.067; Fig. 9-8). Populations occurring in wetlands on the unfarmed coteau showed no relation
between acute stress response and any measure of land use. Acute stress response was unrelated to
any measure of wetland chemistry in either group.
Larvae occurring in wetlands on the drift plain grew faster and reached a larger size by the last
sample date than did larvae on the coteau (Fig. 9-9). In good-condition wetlands, the acute stress
response declined significantly with increasing larval size (Fig. 9-10), while baseline levels increased
a
5000
4500
4000
3500
3000
o 2500
b
•£ 2000
o
£ 1500
h
O 1000
500
Loir Medium High
Acute stress applied
Figure 9-7. Plasma corticosterone levels after low (< 6 min), medium (6-30 min), and high (> 30 min)
acute stress.
169
-------
(Fig. 9-11); no such relation was evident among larvae in poor-condition wetlands (Figs. 9-10 and 9-
11). Similarly, acute stress response was significantly negatively associated with unstressed
corticosterone levels in larvae from good-condition wetlands (Fig. 9-12), but not in those from
poor-condition wetlands.
Laboratory test. Because EPA has discontinued hormone research at ERL-C where the tests
were done, and reassigned personnel who worked on the experiment, data analysis has yet to be
finalized. Inspection of the data revealed that chronic exposure to Guthion did lower brain
cholinesterase, as expected, and thus constituted a chronic stressor. Acute stress consistently resulted
in release of corticosterone, but the magnitude of the acute stress response did not seem to vary
between control and Guthion-exposed larvae. Likewise, larvae exposed for 10 or 20 days did not
appear to have varying responses to acute stress, although 20 days may be too short a time to assess
chronic stress. The concentrations of corticosterone in plasma of larvae in the lab were very similar to
the concentrations of those in the field for both control and acutely stressed animals. More detailed
analyses will be conducted, as time and new work schedules permit.
9.2.5 Evaluation and Recommendations
Tiger salamander response to acute stress varied between good- and poor-condition wetlands.
Those larvae living in good-condition wetlands responded to acute stress in an organized manner, with
the acute stress response declining as the larvae approached the size of metamorphosis; baseline
levels of corticosterone increased as the larvae approached the size of metamorphosis. Larvae in
poor-condition wetlands did not show the typical increase in baseline corticosterone as they approached
metamorphosis, and corticosterone release in response to acute stress showed no pattern with respect
to larval size. Larvae in these poor-condition wetlands tended to show a declining acute stress
response as the amount of cropland surrounding the wetlands increased. These observations suggest
that tiger salamander larvae make physiological accommodations to living under disturbed conditions.
Although such physiological processes are of considerable importance in understanding population-level
response to wetland condition, several issues must be resolved before corticosterone levels can be
used in an operational monitoring program. First, sampling must be expanded, so that wetland condition
is not confounded with subspecific distribution of salamanders. Second, more extensive laboratory
experiments must be carried out, with chronic exposures extended to periods more comparable to larval
residence in wetlands; a variety of common agricultural pesticides, alone and in realistic combinations,
should be used as stressors. Third, other biomarkers, such as white blood cell counts, plasma
170
-------
Table 9-6. Land use at sample wetlands. Site numbers correspond to locations on maps in Appendix
9.2.2. Land use categories are the same as those used in Section 6.0.
Site
Number
2
6
7
8
11
13
14
15
16
17
20
22
23
24
25
26
27
28
31
Condition
Good
Poor
Poor
Poor
Good
Poor
Poor
Good
Good
Poor
Poor
Good
Good
Good
Good
Good
Poor
Poor
Poor
Surrounding land use (%)
Pasture Hay
25
25
50
5
20 25
25
75
65
12
25 25
Idle
100
50
20
100
90
20
75
100
50
100
100
100
100
35
25
33
25
Crop
75
25
30
5
35
25
50
63
66
25
cholinesterase levels, serum glucose concentrations, and possibly shock protein synthesis, should be
considered in concert with both stress and reproductive steroid analysis, to better understand the range
of physiological response to wetland conditions. If connections can be established between these
biomarkers and wetland condition, we will not only be able to use the biomarkers for monitoring, but
also as a step in understanding the mechanisms of population-level responses to environmental
conditions. Such an understanding will give managers a valuable tool in early detection of populations
in danger of decline, and enhance their ability to mitigate in favor of these populations before regulatory
action is required.
171
-------
6000
5000
QO 4000
ft
s^»
0)
o 3000
o 2000
o
o
1000
i i I
* .
0 10 20 30 40 50 60 70 80 90 100
Percent cropland
Figure 9-8. Relation between amount of cropland surrounding wetlands on the drift plain and the acute
stress response.
172
-------
140
^ 180
a
5 100
*j
d
v
4*
I
(1
m 80
60
Drlti plain.
Coteau
2 4 « 8 10
Sampling period (July - August)
12
Figure 9-9. Size of tiger salamander larvae captured at different sampling periods from wetlands
occurring on the drift plain and on the coteau.
173
-------
Poor Condition Wetlands Good Condition Wetland!
r - 0.20 r • -0.65
^, 6000
_j
a
*£ 5000
A
g 4000
o
t.
| 3000
o
o
s 2000
o
o
« 1000
"3
Q 0
X X
*
. a
*
.
,*
•
• * -
*
•
XX
x /
^ 6000
j_4
a
^ 5000
£
§ 4000
Q
h
| 3000
o
o
5 8000
0
o
« 1000
^
» 0
^ ^
' ^ ' • * '
b
-
' -
,
*
> / 1 I • 1
-------
Poor Condition Tetlandi Good Condition Vetlandi
r - 0.01 r - 0.65
4000
S> 3600
0
\ 3000
""' 2500
o
o 8000
* 1600
0
M
% 1000
o &00
0
' '
ss 1 1 1 1 1
" ik ™
V
-
.
• • «•
*
_ ..
^
*
e
. . -
/x i i i i i
4000
O 9600
a
\ aooo
ft,
^ 2600
•
o 8000
•S 1600
o
n
2 1000
,5 600
0
s s
"rS1 — i 1 1
•
"
-
. l^
^^ • '
.s'
^^x^
^^^^ " ••
"^
• Ml
/X I 1 1
fc-a-^-^r
0 BO 90 100 110 1BO 0 80 90 100
Snout-rent length (mm) Snout-rent length (mm)
Figure 9-11. Relation between larval size and baseline plasma corticosterone levels (i.e. levels in larvae
that were note acutely stressed).
175
-------
Poor Condition Vetlanda
r - -0.001
Good Condition V«tlandi
r - -0.68
1
0000
5000
§ 4000
O
| 8000
o
3 8000
o
I 1000 -
I
O '
1000 8000 8000 4000
(pg/ml)
eooo
9000
4000
8000
8000
1000
0
1000 8000 3000 4000
cortloo«t«ron« (pg/ml)
Figure 9-12. Relation between acute stress response and baseline corticosterone levels.
176
-------
Section 10.0
RECOMMENDATIONS FOR CONTINUED USE OF INDICATORS
Lewis M. Cowardin
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
Results from the pilot study suggest a number of changes in design and in the selection of
indicators that would be appropriate for monitoring the condition of prairie wetlands in further studies.
Despite lengthy discussions as to the meaning and definition of condition, no clear definition
emerged during either the planning for the pilot study or its execution. Hughes (1995) discussed
defining acceptable condition. He stressed the need for reference data sets that describe habitat in
good condition. No such data sets exist for the Prairie Pothole Region (PPR). Peterson (1994) stated
that "The biological integrity value, more than any other, assumes the reality of reference conditions,
since it requires that sample site conditions be compared with those of natural wetlands in the region.
Also, because of this requirement biological integrity represents a set of conditions more basic and less
disturbed by human activity (i.e. unmanaged)..." Regional experts at our planning meetings doubted
that such a data set can be constructed because of degradation of most of the prairie potholes,
extreme variation among them, and their vast numbers. Hughes (1995) cautioned against using
reference sites from a random sample: "Similarly, selection of reference sites from randomly selected
sites, especially when those sites are drawn from a population of disturbed sites ... will yield a set of
disturbed reference sites and weak biological criteria."
The problem is that nearly all prairie potholes are disturbed and the degree of disturbance is
confounded with the class of wetland basin and its geological and regional setting. This problem caused
difficulties with the design and execution of the pilot study. We recommend that the definition of
condition relative to reference sites be revisited and a clearly stated plan of action be developed prior to
initiation of any new work by EPA in the Prairie Pothole region.
10.1 SAMPLE FRAME
The 10.4-km2 plots used in the pilot study allowed timely initiation of work, and data derived
from those sample plots did allow us to test selected potential indicators of condition. We found that
177
-------
having both wetlands and uplands mapped prior to selection of wetland basins for sampling and
initiation of field work was essential. In addition, having digital map data (vector files) for each plot
greatly increased efficiency. Without these data it would not have been possible to initiate field work on
schedule. However, there were a number of problems with the plots that preclude their use for further
indicator development and evaluation:
1. The population of plots from which our sample plots was selected is a stratified random
sample with unequal sampling rates in each stratum. The population of plots was
designed for another research purpose.
2. The mapping was out of date and changes have occurred in wetlands since mapping.
In fact, the mapping was accomplished prior to operational mapping by NWI and the
data do not agree with current National Wetland Inventory (NWI) maps. This early
mapping did not receive the quality control given to current mapping and the data
contained topological and classification errors.
3. The plot size is not large enough to assure presence of all classes of wetland basins in
most plots, and among-plot variation in characteristics of the basins in plots was high.
4. There were no data for areas outside the plot boundaries which caused problems when
selecting wetland basins that were bisected by the plot boundary.
We recommend that any future work in the PPR be conducted on 40-km2 hexagon plots that
are part of the EMAP sampling frame and in the selection sequence established for EMAP. Prior to
subsampling or initiation of field work digital wetland data for each hexagon should be prepared as a
subset of standard NWI data, polygons should be collapsed into basins by an agreed upon set of rules,
upland areas should be mapped and digitized according to an agreed upon classification, and all linear
and point features should be buffered so that all features have a measurable area.
10.2 LAND ACCESS
Access to private land was a major problem during the pilot study. Lack of access and the
rescinding of access once granted not only caused us to repeatedly revise our original sampling plan
and design but probably also biased our sample because we suspect that access is granted less
178
-------
frequently on lands that contain poor-condition wetlands than on lands containing good-condition
wetlands. For our purpose of testing indicators, the bias is not that important because our sample was
intentionally selected to represent the best and poorest wetland conditions, not the population of
wetland basins in the PPR. In future work where we are attempting to obtain a statistically valid and
unbiased sample of wetland basins from the PPR the problem will become exceedingly important. In
the pilot a single individual talked personally with each landowner in order to obtain permission and
only verbal permission was requested. In the future, written permission will be required to enter private
land, and personal contact with each landowner is impractical because of the large sample of
hexagons. We suspect that under these conditions the proportion of landowners granting permission
will be much smaller than it was during the pilot study.
Our conclusion is that without purchase of perpetual easements that guarantee access to
sample sites, an unbiased probability sample of wetland basins for ground sampling cannot be obtained
in the PPR. Furthermore, unlike surveys where a second sample can be used to obtain estimates for
the nonresponse bias, there is no practical way to estimate the magnitude of the bias and adjust for it.
10.3 SAMPLE SIZE
Sample size in the pilot was limited by funding rather than by statistical consideration. Given
the exceedingly high spatial and temporal variability of prairie wetlands it is fortunate that we detected
as many differences in condition as we did. We do not believe that statistical difference in condition
class during the pilot should be the sole criterion for acceptance or rejection of potential indicators of
condition. Two things are needed for effective sampling of any indicators selected in the future: First,
we must know the degree of precision required for biological interpretation of condition, and second, we
need some estimate of variance for the indicator to be used. The first requirement is non-statistical and
should be resolved early. The pilot study does furnish some indication of variability to be expected, but
the variance estimates may also be suspect because of the small sample size and the fact that the
populations sampled were at the extremes rather than from the entire population of wetlands.
The preliminary analysis of the sample size problem presented in Section 5 suggests that for
some indicators measured on individual wetland basins variation, both within and among basins, may
be so great that obtaining a biologically meaningful sample may be impractical. We recommend that the
problem of required sample size be addressed prior to the final selection of a suite of indicators to be
included in the next phase of the prairie pothole studies.
179
-------
10.4 SUMMARY OF INDICATOR TESTS
Results obtained from tests of indicators and preliminary recommendations for continuing,
further evaluating, or dropping the indicator in the next phase of the Prairie Pothole Study are
presented in Table 10-1. These recommendations assume that we will continue to sample at selected
wetland basins despite the problems of access and sample size discussed above. During the course of
the studies various investigators have suggested potential new indicators that might be preferable to
those selected for the pilot study. For example, many of the poor condition plots had deltas of silt that
had eroded from the upland; thus, one investigator suggested that these deltas might be delineated on
photographs and their presence used as a landscape indicator of condition. The individual sections of
this report suggest other potential new indicators. The Northern Prairie Science Center (NPSC) is in the
process of soliciting suggestions for additional new indicators. We recommend that these new indicators
as well as some of those used in the pilot be considered for inclusion in the next phase of the studies.
10.5 RECOMMENDED NEW APPROACH
We recommend a new approach to EMAP sampling for the PPR. Our recommendation is
based on biological, statistical, and practical considerations. The new approach is that an individual
wetland basin should not be considered as a sampling unit. Rather, a group of wetlands, frequently
referred to as a wetland complex, is the entity whose condition would be evaluated. We define a
complex as all of the wetland basins and their associated uplands within a 40-km2 hexagon. The
hexagon then becomes the sampling unit and all indicators would be indicators of landscape condition.
Landscapes as sampling units make biological sense. All of the experts who participated in the
planning phase of the pilot stressed that condition of prairie potholes must be evaluated in terms of
complexes that include both the wetland basins and their associated uplands. Many of the candidate
biological indicators of condition, especially birds and mammals, move freely among the wetland basins
and their associated uplands. For example, upland nesting ducks require a complex of wetlands that
includes temporary, seasonal, and semipermanent wetland basins as well as safe nesting cover on the
adjacent uplands. Where we find high populations and good production of ducks, these landscape
components are present. A reference landscape should have these components present. An impaired
landscape may have lost one or more of them; therefore, lack of duck abundance as a potential
indicator will reflect the loss of condition.
180
-------
Table 10-1. Summary of recommendations for Indicator measurements tested during a pilot study of indicators of wetland condition in the
prairie pothole region.
Section
Measurement
Recommendation
Notes
Landscape
oo
Number and area of
basins
Distance between
basins and shoreline
Shoreline
development index
Drained wetland
basins
Length of drainage
ditch
Wetland change index
Area of cropland
Area of exposed soil
Estimated breeding
ducks
Continue
Evaluate
Evaluate
Continue
Continue
Continue
Continue
Continue
Revise
These measurements should be continued to
furnish baseline data for other variables, but not as an
indicator of condition.
Not an indicator of condition, but may be
valuable to evaluation of other indicators.
Not an indicator of condition, but may be
valuable to evaluation of other indicators.
Good indicator of condition of wetlands in
landscape. Easy from remote sensing.
Good indicator of condition of wetlands in
landscape. Easy from remote sensing.
Drop the April-May period because of ice and snow causing
interpretation errors. June-July period appears best from
limited pilot data.
Simplest and most promising direct measurement of condition.
Need further studies linking biological indicators and cropland
abundance.
Good indicator of condition. Needs further study
and expanded development of remote sensing or
supplemental data sources to estimate crop types.
Appears to be a good indicator. Counts of ducks should be
made directly on plots to improve estimates of 7.
-------
Table 10-1. (continued)
Section
Measurement
Recommendation
Notes
Macro-
invertebrates
Siltation
Water depth
Invertebrates
Sediment Traps
Water-level recorders
Drop
Evaluate
Continue
Water quality
Turbidity, EC, pH,
Drop
Soils
Sedimentation
Soil composition
Evaluate
Modify
High within and among wetland basin variation precludes use
of invertebrate remains as a practical indicator of condition.
Future work should evaluate invertebrates as an indicator at
the landscape scale.
This indicator did not detect differences in condition when
traps were placed at the center of the wetland basin. Suggest
further work to evaluate alternative placement of traps.
The recorders reported in Section 10.1 may furnish
information that will evaluate hydrologic condition. In addition,
these devices may furnish data to help evaluate results
obtained in Section 4.0. Commercial devices tested are too
expensive to be useful.
These indicators were dropped during the study because of
the need for wetland basins containing water. The high
degree of variation in water permanence and in water quality
among basins make water quality measurements impractical
in the PPR.
C-137 showed some promise for evaluating sedimentation;
however, large samples would be required and costs are high.
Phosphorous (at 0-15 cm depth) and organic matter in
sediments were the only constituents showing promise as
indicators of wetland condition.
-------
Table 10-1. (continued)
Section
Measurement
Recommendation
Notes
Vegetation
Pesticides
Salamanders
Amount of
unvegetated bottom
Standing dead
vegetation
Percent open water
Taxa richness
Native perennial
Immunosorbent assay
Populations
Hormonal response
Continue
Drop
Drop
Continue
Continue
Continue
Dropped
Evaluate
Measure the amount of unvegetated bottom in the wet
meadow zones of wetland basins.
No condition differences.
No condition differences.
Good indicator for use in the wet meadow zone.
Good indicator for wet meadow zone. Adjust for plant
community size.
Tests for atrazine distinguished condition, but
should be related to corn growing areas. Recommend
addition of kits appropriate for other crops and groups of
agricultural chemicals.
These vertebrates are confined to specific wetland classes
that are to sparse to be sampled in the EMAP sampling
frame.
Results were encouraging but use will require extended
sampling, additional laboratory tests, and consideration of
other biomarkers.
-------
Landscapes as a sampling unit make statistical sense. The wetland values suggested for prairie
potholes, biological integrity, harvestable productivity, water quality improvement, and flood attenuation
(Peterson, 1994), can be measured at the landscape rather than an individual basin scale. Much of the
variation that we encountered in the pilot studies was among wetland basins. If we sample wetland
complexes much of this variation might average out. Furthermore many of the measurements that are
specific to individual basins proved impractical or did not distinguish between good and poor condition.
Stressors tend to affect entire landscapes rather than individual wetland basins. For example, individual
wetland basins are seldom drained; rather, whole complexes are drained. Runoff of agricultural
chemicals into basins impairs individual basins as well as the basins in the crop field being treated. We
suggest that we can obtain data for any hexagon by (1) restricting the indicators to those that can be
monitored remotely, (2) by accepting indices obtained from roadside transects as representing condition
of the hexagon or (3) using models that relate biological indicators to attributes that can be measured
by remote sensing. In our opinion this is a legitimate strategy as long as we are not attempting to
extrapolate the roadside estimate to the entire hexagon. For example, we would not attempt to estimate
the number of ducks on the hexagon. We only assume that more ducks would be seen from roads in
good condition landscapes than in poor condition landscapes.
Landscapes as sampling units make practical sense. We have shown that a valid probability
sample of wetland basins is not possible in the PPR. Using the hexagon, with measurements made
either remotely or as indices derived from roads or point samples avoids the problem of access. We
also believe that these rapid survey techniques lend themselves to survey of more sampling units at
less cost.
We also recommend increased effort on studies not directly related to the probability sample,
but rather designed to develop relations between biological indicators and landscape features that can
be determined remotely. These studies should be followed by development of appropriate models. Such
studies need not be conducted on randomly selected areas. For example, if we determine that
fragmentation of habitat causes a decrease in a biological indicator we can use remote sensing to
determine if fragmentation is increasing or decreasing on the habitats and then predict that the indicator
species is increasing or decreasing. Studies that relate biological indicators to measurable
environmental variables are also essential to interpreting results derived from EMAP sampling. The
process must be understood before EMAP results can be translated into remedial action. If a species or
group of species that indicates environmental condition is shown to be in decline through EMAP
monitoring, the process that causes the decline must be understood prior to management to reverse
the decline and restore good condition.
184
-------
Section 11. Q
PERSONNEL AND COSTS
Lewis
U,§,
Northern Prairie Science Center
Jamestown, North Da
The nature of the work conducted in the pilot study is technically complex and requires
expertise in a number of fields. That §xp§rtj§e i§ expensive. Our summary refers to the pilot study
where many of the techniques and procedures were developed specifically for the various studies. We
suspect that when final indicators are selected some of the measurements might be made more
cheaply on an operational basis.
11.1 EXPERTISE REQUIREMENTS
The project required considerable coordination between the various projects and EPA, the
funding organization. A high degree of familiarity with thf Environmental Monitoring and Assessment
Program (EMAP) as well as the ecology of the Prairie Pothole Region (PPR) was required to
accomplish this coordination. The project also had major requirements for study plan development,
review of study plans, preparation of reports, and quality assurance. These activities required the same
expertise as coordination.
Competent administrative support was required to make sure that field work moved ahead on
schedule and that conflicts among the various components of the study did not occur. The primary task
of administering the project was obtaining landowner access to the study areas. Administration of the
project did not require specific expertise in a particular field but did require good knowledge of the area
and skill in personal relations with landowners.
Many of the data were gathered by remote sensing and processed by GIS computer
techniques. This work required an individual with general knowledge of remote sensing and specific
knowledge of the ecology of the PPR. Experience with the maps and other products produced by the
National Wetlands Inventory (NWI) as well as experience in planning aerial video\photographic mission
185
-------
was also required. The work required technical support and a high degree of expertise with computer
processing and the software being used for interpretation and data analysis.
Planning, conducting, analyzing, and reporting the information of the project required individuals
with expertise in specialized fields. These included vertebrate ecology, waterfowl ecology, plant ecology
and taxonomy, invertebrate biology, limnology, soil science, and statistics. When the EMAP sampling
becomes operational, finding the appropriate pool of individuals with the required expertise will present
a problem. Although well trained technicians could help with much of the work specialized expertise will
also be necessary. Using graduate students at the PhD level and under the guidance of experienced
researchers might furnish one solution. In our opinion, the type of indicator measurements made in the
pilot study require a multidisciplinary team approach. The technical team requires efficient logistical and
administrative support to conduct the field work in a short time and over an extensive area.
11.2 COSTS
Costs for the various pilot project studies in calendar years 1992, 1993, and 1994 are
summarized in Table 11-1. These do not include the costs for planning the project and preparing study
plans incurred in the fall of 1991. These costs (Table 11-1) do include a considerable contribution by
the Fish and Wildlife Service (FWS), however, which would not be available for operational work. For
planning purposes, an approximate cost per sample unit can be obtained by dividing the project's cost
by the number of years and number of plots per year. For example, the total cost for landscape
variables was $253,000 for 3 years and 16 plots, yielding a per-plot cost of $5,271 per year. These
costs are probably high because as the work progressed our efficiency increased in a number of areas.
In addition, a number of tasks could probably be performed by temporary staff at a lower pay rate than
in the pilot.
186
-------
Table 11-1. Costs of the Prairie Pothole Pilot Project by activity. Costs are for calendar years 1992,1993, and 1994.
Salaries
Permanent
Project
Administration
Landscape
Macroinvertebrates
Vegetation
Sampling device
Soils
Hormonal Response
Total
Salaries
$1000
92.53
130.06
49.33
133.11
56.07
6.87
36.86
504.83
FTE
1.46
2.54
0.88
2.35
0.99
0.12
0.85
9.19
Temporary
Salaries
$1000
8.01
27.50
85.15
0.00
61.78
0.00
12.26
194.70
FTE
0.39
1.21
3.64
0.00
2.68
0.00
0.59
8.51
Vehicles
$1000
3.07
27.65
37.80
3.35
33.00
0.00
4.08
24.94
Other3
$1000
8.49
27.65
37.80
3.35
33.00
0.00
17.60
127.89
Operating Costs
Contracts
$1000
0.00
0.00
0.00
0.00
0.00
90.64
0.00
90.64
Overhead3
$1000
36.99
62.91
58.46
45.89
51.40
32.18
23.36
311.19
Total
$1000
149.09
253.54
235.62
184.94
207.15
129.69
94.16
1254.19
"Includes equipment, supplies, travel, and aircraft.
'Includes administration, statistical support, etc.
-------
REFERENCES
American Society of Ichthyologists and Herpetologists, The Herpetologists' League, and Society for the
Study of Amphibians and Reptiles. 1987. Guidelines for use of live amphibians and reptiles in
field research.
Andraski, B.J., D.H. Mueller, and T.C. Daniel. 1985. Phosphorus losses in runoff as affected by tillage
Soil Sci. Soc. Am. J. 1523-1527.
Andresen, H., J.P. Bakker, M. Brongers, B. Heydemann, and U. Irmler. 1990. Long-term changes of
salt marsh communities by cattle grazing. Vegetetio 89:137-148.
Bakker, J.P. 1985. The impact of grazing on plant communities, plant populations and soil conditions on
salt marshes. Vegetatio 62:391 -398.
Barker, W.T. and G.W. Fulton. 1979. Analysis of wetland vegetation on selected areas in southwestern
North Dakota. N.D. Regional Environmental Assessment Program. Rep. No. 79-15. North
Dakota State University. Fargo. 132 pp.
Bazely, D.R., and R.L. Jeffries. 1986. Changes in composition and standing crop of saltmarsh
communities in response to the removal of a grazer. J. Ecol. 74:693-706.
Beeftink, W.G. 1977. The structure of salt marsh communities in relation to environmental disturbances.
pp. 77-93 in (R.L. Jefferies and A.J. Davy eds.) Ecological processes in coastal environments.
Blackwell Scientific Publications, Oxford.
Belton, P. and R.H. Kempster. 1963. Some factors affecting the catches of Lepidoptera in light traps.
Can. Ent. 95:832-837.
Belton, P. and A. Pucat. 1967. A comparison of different lights in traps for Culicoides (Diptera:
Ceratopogonidae). Can. Ent. 99:267-272.
Bergelson, J. 1990. Spatial patterning in plants: opposing effects of herbivory and competition. J. Ecol.
78:937-948.
Bernard, J.M. 1974. Seasonal changes in standing crop and primary production in a sedge wetland and
an adjacent dry old-field in central Minnesota. Ecology 55:350-359.
Biondini, M.E., P.W. Mielke Jr., and K.J. Berry. 1988. Data-dependent permutation techniques for the
analysis of ecological data. Vegetatio 75:161 -168.
Bloesch, J., and N.M. Burns. 1980. A critical review of sedimentation trap techniques. Schweiz. Z
Hydrol. 42:15-55.
Blomqvist, S., and L. Hakanson. 1981. A review on sediment traps in aquatic environments. Arch.
Hydrobiol. 91:101-132.
Bluemle, J.P. 1991. The face of North Dakota. North Dakota Geological Survey, Educational Series 21.
177pp [map].
189
-------
Borror, D.J., D.M. DeLong, and C.A. Triplehorn. 1981. An introduction to the study of insects. 5th
edition. Saunders College Publ. 827pp.
Brown, T.L. 1974. New York landowners' attitudes toward recreation activities. Trans, of the N. Am.
Wildl. Nat. Res. Conf. 39:255-263.
Brown, T.L., D.J. Decker, and J.W. Kelley. 1984. Access to private lands for hunting in New York:
1963-1980. Wildl. Soc. Bull. 12:344-349.
Cahoon, D.R., and R.E. Turner. 1989. Accretion and canal impacts in a rapidly subsiding wetland II.
Feldspar marker horizon technique. Estuaries 12:260-268.
Campbell, P.M., T.G. Pottinger, and J.P. Sumpter. 1994. Preliminary evidence that chronic confinement
stress reduces the quality of gametes produced by brown and rainbow trout. Aquaculture
120:151-169.
Cole, G.A. 1983. Text book of limnology. Waveland Press. Prospect heights, II. 3rd Ed.
Conover, W.J. 1980. Practical nonparametric statistics, 2nd edition. New York, NY: John Wiley and
Sons, Inc. 493 pp.
Cowardin, L.M. 1982. Some conceptual and semantic problems in wetland classification and inventory.
Wildl. Soc. Bull. 10:57-60.
Cowardin, L.M., D.H. Johnson, T.L. Shaffer, and D.W. Sparling. 1988. Applications of a simulation
model to decisions in mallard management. U.S. Fish Wildl. Serv. Tech. Rep. 17. 28pp.
Cowardin, L.M., V. Carter, F.C. Golet, and E.T. LaRoe. 1979. Classification of wetlands and deepwater
habitats of the United States. U.S. Fish Wildl. Serv. FWS/OBS-79/31. 103pp.
Cowardin, L.M., T.L. Shaffer, and P.M. Arnold. 1995. Evaluations of duck habitat and estimation of
duck population sizes with a remote-sensing-based system. U.S. Dept. Int. Nat. Biol. Surv.
Biol. Sci. Rep. 2. 26pp.
Crumpton, W.G. 1989. Ch.6 algae in northern prairie wetlands Neely, R.K. and J.L. Baker, 1989. Pp
188-203. In A. van der Valk (ed.). Northern Prairie Wetlands. Iowa State Univ. Press,
Ames, IA.
Dahnke, W.C. (ed) 1988. Recommended chemical soil test procedures for the North Central Region,
Publ. No. 221 (revised). N. D. Ag. Exp. St., North Dakota State Univ., Fargo.
Dahnke, W.C. and D.A. Whitney. 1988. Measurement of salinity, pp 32-34. In: W.C. Dahnke, (ed)
Recommended chemical soil test procedures for the North Central Region. North Central
Region Publication no. 221 (revised). N. D. Agr. Exp. St., North Dakota State Univ., Fargo.
Dauphin-Villemant, C. and F. Xavier. 1987. Nychthemeral variations of plasma corticosteroids in
captive female Lacerta vivipara Jacquin: influence of stress and reproductive state. Gen.
Comp. Endocrinol., 67:292-302.
Davidson, A, J.E. Gallagher, and H.S. Hsiao. 1973. Reactions of some moths and aquatic insects to
light traps fitted with polarizing filters. Ann. Entomol. Soc. Am. 66:232-233.
190
-------
Day, P.R. 1965. Particle fractionation and particle size analysis, pp. 535-540 in C.A. Black (ed.)
Methods of soil analysis, Part 1. Agronomy Mongr. 9. American Society of Agronomy,
Madison, Wl.
DeJong, E., C. Wang, and H.W. Rees. 1986. Soil redistribution on three cultivated New Brunswick
hillslopes calculated from Cs-137 measurements, solum data, and the USLE. Can J Soil Sci
66:721-730.
DeLaune, R.D., W.H. Patrick, Jr., and R.J. Buresh. 1978. Sedimentation rates determined by 137Cs
dating in a rapidly accreting salt marsh. Nature 274:532-533.
Dupont, W., P. Bourgeois, A. Reinberg, and R. Vaillant. 1979. Circannual and circadian rhythms in the
concentration of corticosterone in the plasma of the edible frog (Rana esculenta L). J. Endocr.,
80:117-125.
Dwire, K.A. 1994. Research and quality assurance methods used in the EMAP Prairie Wetlands Pilot
Study. EPA/620/R-94/018. U.S. Environmental Protection Agency, Office of Research and
Development, Environmental Research Laboratory and the Environmental Monitoring and
Assessment Program, Corvallis, Oregon.
Eckert, D.J. 1988. Recommended pH and lime requirement tests, pp 6-8. In: W.C. Dahnke, (ed)
Recommended chemical soil test procedures for the North Central Region. North Central
Region, Publ. No. 221 (revised). N. D. Agr. Exp. St., North Dakota State Univ., Fargo.
Fellows, D.P., and T.K. Buhl. 1995. Research access to privately owned wetland basins in the Prairie
Pothole Region of the U.S. Wetlands (15)4:330-335.
Gardner, W.D. 1980. Field assessment of sediment traps. J. Mar. Res. 38:41-52.
Geller, M.D., and J.J. Christian. 1982. Population dynamics, adrenocortical function, and pathology in
Microtus pennsylvanicus. J. Mamm., 63:85-95.
Gleason, Robert A. 1996. Influence of agricultural practices on sedimentation rates, aquatic
invertebrates, and bird.use in prairie wetlands. M.S. Thesis. Humbolt State University.
Arcata, CA 92 p.
Grubb, P.J. 1977. The maintenance of species-richness in plant communities: the importance of the
regeneration niche. Biol. Rev. 52:107-145.
Grue, C.E., M.W. Tome, T.A. Messmer, D.B. Henry, G.A. Swanson, and L.R. DeWeese. 1986.
Agricultural chemicals and prairie pothole wetlands: meeting the needs of the resource and the
farmer - U.S. perspective. Trans. North Am. Wildl. Nat. Resour. Conf. 54:43-58.
Hammond, M.C. 1961. Habitat improvement studies at Lower Souris National Wildlife Refuge-past,
present and proposed. U.S. Fish and Wildl. Serv. 8pp.
Harding, W.C., J.G. Hartsock, and G.G. Rohwer. 1966. Blacklight trap standards for general insect
surveys. Bull. Entomol. Soc. Am. 12:31-32.
Hargrave, B.T., and N.M. Burns. 1979. Assessment of sediment trap collection efficiency. Limnol.
Oceanogr. 24:1124-1136.
191
-------
Harlow, H.J., E.T. Thome, E.S. Williams, E.L. Belden, and W.A. Gern. 1987. Cardiac frequency: a
potential predictor of blood cortisol levels during acute and chronic stress exposure in Rocky
Mountain bighorn sheep (Ovis canadensis canadensis). Can. J. Zoo/., 65:2028-2034.
Hellings, S.E., and J.L. Gallagher. 1992. The effects of salinity and flooding on Phragmites australis.
J.Appl. Eco/. 29:41-49.
Hontela, A., J.B. Rasmussen, C. Audet, and G. Chevalier. 1992. Impaired cortisol stress response in
fish from environments polluted by PAHs, PCBs, and mercury. Arch. Environ. Contam. Toxicol.
22:278-283.
Hughes, R.M. 1995. Defining acceptable biological status by comparing with reference conditions. In T.
Simon and W. Davis eds. Biological assessment and criteria: tools for water resource planning
and decision making. Lewis Publ., Chelsea Ml.
Johnson, W.C., T.L. Sharik, R.A. Mayes, and E.P. Smith. 1985. Nature and cause of zonation
discreteness around glacial prairie marshes. Can. J. Bot. 65:1622-1632.
Johnson, D.H., D.W. Sparling and L.M. Cowardin. 1987. A model of the productivity of the mallard
duck. Eco/. Model 38:257-275.
Johnson, R.A., and D.W. Wichern. 1988. Applied multivariate statistical analysis, 2nd edition.
Englewood Cliffs, NJ: Prentice Hall. 607 pp.
Jones, W.L., and W.C. Lehman. 1987. Phragmites control and revegetation following aerial
applications of glyphosate in Delaware, pp. 185-196 in W. R. Whitman and W. H. Meredith,
eds. Waterfowl and wetlands symposium: Proceed, of a Symp. on Waterfowl Wetlands
Management in Coastal Zone of the Atlantic Flyway. Delaware Coastal Manage. Prog.,
Delaware Dept. Nat. Res. Envir. Contl., Dover.
Kantrud, H.A., and R.E. Stewart. 1984. Ecological distribution and crude density of breeding birds on
prairie wetlands. J. Wildl. Manage. 48:426-437.
Kantrud, H.A., G.L. Krapu, and G.A. Swanson. 1989. Prairie Basin Wetlands of the Dakotas: A
community Profile. U.S. Fish Wild. Setv. Biol. Rept. 85(7.28). 111pp.
Karr, J.R., and D.R. Dudley. 1981. Ecological perspective on water quality goals. Environmental
Management 5:55-68.
Kiefer, J.L. 1985. Attitudes of landowners and landusers toward recreation on private land in
Sheboygan County, Wisconsin. M. S. Thesis, Univ. of Wisconsin, Stevens Point. 80pp.
Kirkpatrick, J.F., C.B. Baker, J.W. Turner Jr., R.M. Kenney and V.K. Ganjam. 1979. Plasma
corticosteroids as an index of stress in captive feral horses. J. Wildl. Manage. 43:801-804.
Kitts, J.R. and J.B. Low. 1974. Utah landholders' attitudes toward hunting. Trans. N. Am. Wildl. and
Nat. Resour. Conf. 39:180-185.
Kling, H.J. 1975. Phytoplankton successions and species distribution in prairie ponds of the Erickson-
Elphinstone District, southwestern Manitoba. Fish. Mar. Serv. Tech. Rep. No. 512,
Winnipeg. 31 p.
192
-------
Knudsen, D. and D. Beegle. 1988. Recommended phosphorus tests, pp 12-15. In: W.C. Dahnke, (ed)
Recommended chemical soil test procedures for the North Central Region. North Central
Region, Publ. No. 221 (revised). N. D. Agr. Exp. St., North Dakota State Univ., Fargo.
Koeln, G.T., P. Caldwell, D.E. Wesley and J.E. Jacobson. 1986. Inventory of wetland with Landsat's
thematic mapper. Proc. Can. Symp. Remote Sens. 10:153-162.
Lau, Y.L. 1979. Laboratory study of cylindrical sedimentation traps. J. Fish Res. Board. Can.
36:1288-1291.
Leibowitz, N.C., L. Squires, and J.P. Baker. 1991. Research plan for monitoring wetland ecosystems.
U.S. Envirn. Prof. Agency. EPA/600/3-91/010. 157pp.
Licht, P., B.R. McCreery, R. Barnes and R. Pang. 1983. Seasonal and stress related changes in
plasma gonadotropins, sex steroids, and corticosterone in the Bullfrog, Rana catesbeiana.
Gen. Comp. Endocrinol., 50:124-145.
Loveland, T.R., J.W. Merchant, D.O. Ohlen, and J.F. Brown. 1991. Development of a land-cover
characteristics database for the conterminous United States. Photogr. Eng. Remote Sens.
57:1453-1463.
Mann, G.E. 1974. The Prairie Pothole Region~a zone of environmental opportunity. Naturalist
25(4):2-7.
Mason, W.T., Jr. and J.E. Sublette. 1971. Collecting Ohio River Chironomidae (Diptera) with a floating
sticky trap. Can. Ent. 103:397-404.
McDonald, I.R. and M.J. Tartt. 1982. Steroid hormones in the blood plasma of Townsend's vole
(Microtus townsendir). Can. J. Zoo/. 60:2264-2269.
McDonald, I.R., A.K. Lee, K.A. Thank and R.W. Martin. 1988. Concentration of free glucocorticoids in
plasma and mortality in the Australian bush rat. J. Mamm. 69:740-748.
McNaughton, S.J. 1979. Grazing as an optimization process: grass-ungulate relationships in the
Serengeti. Am. Nat. 113:691-703.
McNaughton, S.J. 1986. On plants and herbivores. Am. Nat. 128:765-770.
-Milliken, G.A. 1990. Analysis of covariance: repeated measures and split plot designs. Proc. SAS
Users Group Intern. Conf. 15:1268-1276.
Milliken, G.A. and D.E. Johnson. 1984. Analysis of messy data, Vol.l: designed experiments. Van
Nostrand Reinhold Co., New York, NY. 473pp.
Mitch, W.J., and J.G. Gosselink. 1986. Wetlands. Van Nostrand Reinhold, Co., New York. 539 pp.
Moore, F.L. and L.J. Miller. 1984. Stress-induced inhibition of sexual behavior: corticosterone inhibits
courtship behaviors of a male amphibian (Taricha granulosa). Hormones and Behavior
18:400-410.
193
-------
Moore, F.L. and P. Deviche. 1987. Neuroendocrine processing of environmental information in
amphibians, p. 19-45. In: M. H. Stetson (ed.). Process, of Envir. Infor. in Vertebrates.
Springer-Verlag, New York.
Moore, M.C., C.W. Thompson and C.A. Marler. 1991. Reciprocal changes in corticosterone and
testosterone levels following acute and chronic handling stress in the Tree Lizard, Urosaurus
ornatus. Gen. Comp. Endocrinol. 81:217-226.
Moore, F.L. 1983. Behavioral endocrinology of amphibian reproduction. BioScience 33:557-561.
Nash, R.G. 1988. Chapter 5. Dissipation from soil, p. 131-169. /n:R. G rover (ed.). Envir. Chem.
Herbicides, CRC Press, Inc., Boca Raton, Florida.
Neely, R.K. and J.L. Baker. 1989. Chapter 4. Nitrogen and phosphorus dynamics and the fate of
agricultural runoff, pp 92-131. In A. Van der Valk (ed.). Northern Prairie Wetlands. Iowa State
Univ. Press, Ames.
Olsen, S.R., C.V. Cole, F.S. Watenabe, and L.A. Dean. 1954. Estimation of available phosphorus in
soils by extraction with sodium bicarbonate. USDA Circular 939:1 -19.
Orchinik, M., P. Licht and D. Crews. 1988. Plasma steroid concentrations change in response to
sexual behavior in Bufo marinus. Hormones and Behavior 18:338-350.
Overton, W.S., White, and D.L. Stevens Jr. 1990. Design report for EMAP environmental monitoring
and assessment program. U.S. Envir. Protect. Agency. EPA/600/3-91/053. Corvallis, Oregon.
Parsons, P.A. 1990. The metabolic cost of multiple environmental stresses: implications for climatic
change and conservation. Trends Ecol. Evol. 5:315-317.
Paulsen, S.G., N.S. Urquhart, and D.P. Larsen. 1993. Prototype annual report. Pages 157-176 in D. P.
Larsen and S. J. Cristie [eds.] EMAP-Surface waters pilot report. U.S. Envirn. Protect Agency.
EPA/620/R-93/003. Washington DC.
Peterson, S.A. 1994. The Environmental Monitoring and Assessment Program (EMAP): Its objectives,
approach and status relative to wetlands. In Aubrect, G., G.Dick, and C. Prentice, (eds.).
Monitoring of ecological change in wetlands of Middle Europe. Proceed, of an international
workshop. Linz, Austria. Stapfia 31, Linz, Austria, and IWRB Special Publ. No. 30. Slimbridge,
UK. 224pp.
Pywell, H.R., and H.A. Niedzweadek. 1980. The wetlands analytic mapping system - WAMS. Pp. 261-
270 in Analytical Plotter Symposium. Am. Soc. Photogrametry. Reston, VA.
Reed, P.B. Jr. 1988. National list of plant species that occur in wetlands: North plains (Region 4). U.S.
Fish Wildl. Serv. Biol. Rep. 88(26.4).
Richardson, J.L. and J.L. Arndt. 1989. What use prairie potholes? J. Soil Water Conserv.
44(3): 196-198.
Ritchie, J.C., J.R. McHenry, and A.C. Gill. 1973. Dating recent reservoir sediments. Limnol. Oceanogr.
18:254-263.
194
-------
Rosen B.H., P. Adamus, and H. Lai. 1995. A conceptual model for the assessment of depressional
wetlands in the prairie pothole region. J. Wetl. Manage. Ecol. 3:195-208.
Sargeant, A.B. and D.G. Raveling. 1992. Mortality during the breeding season. Pages 396-422. In
Batt, B.D.J., A. D. Afton, M.G. Anderson, C.D. Ankney, D.H. Johnson, and G.L. Krapu eds.
The Ecology and management of breeding waterfowl. Univ. Minnesota Press. Minneapolis.
Sargeant, A.B., R.J. Greenwood, M.A. Sovada, and T.L. Shaffer. 1993. Distribution and abundance of
predators that affect duck production - The prairie pothole region. U.S. Fish Wildl. Serv., Fish
Wildl. Resour. Pub. 194. 96pp.
SAS Institute, Inc. 1992. SAS Technical Report p-229, SAS/STAT Software: Changes and
Enhancements, Release 6.07. SAS Institute, Inc., Gary, NC. 620pp.
SAS Institute Inc. 1988. SAS/Stat User's Guide, Release 6.03 Edition. SAS Institute, Gary, NC, USA.
SAS Institute, Inc. 1989. SAS/STAT User's Guide, Version 6, 4th ed., Vol.2. SAS Institute, Inc.
Gary, NC. 846p.
Schaeffer, D.J., E. E. Herricks, and H. W. Kerster. 1988. Ecosystem Health: 1. Measuring ecosystem
health. Envir. Manage. 12:(4)445-455.
Schindler, D.W. 1977. Evolution of phosphorus limitation in lakes. Science 195:260-262.
Schulte, E.E. 1988. Recommended soil organic matter tests, pp 29-32. In: W.C. Dahnke, (ed)
Recommended chemical soil test procedures for the North Central Region. North Central
Region Publication no. 221 (revised). N. D. Ag. Exp. St., North Dakota State Univ., Fargo.
Schultz, B.D. 1987. Biotic responses of Typ/ja-monodominant semipermanent wetlands to cattle
grazing. M. S. Thesis, South Dakota State Univ., Brookings. 92pp.
Seal, U.S. and R.L. Hoskinson. 1978. Metabolic indicators of habitat condition and stress in
pronghorns. J. Wildl. Manage. 42:755-763.
Shaffer, T.L., and W.E. Newton. Duck nest success in the prairie pothole region, 1966-1989. J. Wildl.
Manage, (in press)
Shaffer, T.L. and W.E. Newton. 1995. Duck nest success in the Prairie Potholes. In E. T. Laroe III, G.
S. Farris, C. E. Puckett, and P. D. Doran, editors. Our living resources: a report to the Nation
on the distribution, abundance and health of U.S. plants, animals, and ecosystems. U.S. Dept.
of the Int., Nat. Biol. Serv., Washington, DC.
Sharpley, A.N. and S.J. Smith. 1983. Distribution of phosphorus forms in virgin and cultivated soils and
the potential erosion losses. Soil Sci. Soc. Am. J. 47:581-586.
Sidle J.G. 1983. Ducks, government and private land. Naturalist 34:4 12-19.
Sidle, J.G., and J.W. Ziewitz. 1990. Use of aerial videography in wildlife studies. Wild. Soc. Bull.
18:56-62
195
-------
Smeins, F.E. 1967. The wetland vegetation of the Red River Valley and drift prairie regions of
Minnesota, North Dakota and Manitoba. Ph.D Diss., Univ. of Saskatchewan, Saskatoon.
226pp.
Smith, R.S., and S.P. Rushton. 1994. The effects of grazing management on the vegetation of
mesotrophic (meadow) grassland in northern England. J. Appl. Ecol. 31:13-24.
Smith, A.G. and J.H. Stoudt. 1964. Prairie Potholes. Pages 39-50 in Linduska, J. P. [ed.] Waterfowl
Tommorow. U.S. Dept. Interior. Bur. Sport Fish, and Wildl., Washington, DC.
Smutko, L.S., J.A. Leitch, I.E. Danielson, and R.K. Stroh. 1984. Landowner attitudes toward wetland
preservation policies in the Prairie Pothole Region. N. D. Ag. Exp. St. Ag. Econ. Misc. Rep. 78,
North Dakota State Univ., Fargo. 19pp.
Soil Survey Staff. 1992. Keys to soil taxonomy. SMSS Technical Monog. No. 19, 5th ed. Pocahontas
Press, Blacksburg, VA.
Soileau, J.M., B.F. Hajek, and J.T. Touchton. 1990. Soil Erosion and deposition evidence in a small
watershed using fallout Cesium-137. So/7 Sci. Soc. Am. J. 54:1712-1719.
Steel R. G. D. and J. H. Torrie. 1980. Principles and procedures of statistics, a biometrical approach,
second edition. McGraw-Hill Book Company. New York, NY. 235pp.
Stewart, R. E. 1975. Breeding Birds of North Dakota. Tri-County Center for Environmental Studies.
Fargo, ND. 295pp.
Stewart, R.E. and H.A. Kantrud. 1971. Classification of natural ponds and lakes in the glaciated prairie
region. U.S. Fish Wildl. Serv. Prof. Pap. 585-D. 36pp.
Stewart, R. E., and H. A. Kantrud. 1972. Vegetation of prairie potholes, North Dakota in relation to
quality of water and other environmental factors. U.S. Geol. Surv. Prof. Pap. 585-D. 36pp.
Stoddard, C. H. and A. M. Day. 1969. Private lands for public recreation: is there a solution? Trans. N.
Am. Wildl. and Nat. Resour. Conf. 34:186-196.
Swanson, G. A. 1987. An introduction to the Cottonwood Lake area. Proc. North Dakota Acad. Sci.
41:25.
Thurman, E.M., D.A. Goolsby, M.T. Meyer, M.S. Mills, M.L. Pomes and D.W. Kolpin. 1992. A
reconnaissance study of herbicides and their metabolites in surface water of the midwestern
United States using immunoassay and gas chromatography/mass spectrometry. Envir. Sci.
Technol. 26:2440-2447.
Tome, M.W., C.E. Grue and L. DeWeese. 1991. Ethyl parathion in wetlands following aerial application
to sunflowers in North Dakota. Wildl. Soc. Bull. 19:450-457.
Tome, M.W., C.E. Grue, S. Borthwick, and G.A. Swanson. 1990. Effects of an aerial application of
ethyl parathion on selected aquatic invertebrate populations inhabiting prairie pothole wetlands.
Pages 31-33 in Proc. Env. Conf. on Biota. Northern Great Plains. Bismarck, North Dakota.
196
-------
Turner, M.M.G. 1985. Ecological effects of multiple perturbations on a Georgia salt marsh. Ph.D.
Diss., Univ. of Georgia, Athens. 192pp.
United States Department of Agriculture. 1982. National List of Scientific Plant Names Soil Consent
Serv. SCS-TP-159.
United States Environmental Protection Agency. 1993. EMAP Project Descriptions. EPA/620/R-93/009.
Office of Research and Development, Washington, DC 20460. 106 pp.
van der Valk, A. 1989. Northern prairie wetlands. Iowa State Univ. Press, Ames. 400pp.
Vendrell, P.P. and Zupacic. 1990. Determination of soil nitrate by transnitration of salicylic acid.
Commun. Soil Sci. Plant Anal., 21(13-16):! 705-1713.
Velleman, P.P. and D.C Hoaglin. 1981. Application, basics, and computing of exploratory data analysis.
Duxbury Press. Boston.
Vermeer, J.G., and F. Berendse. 1983. The relationship between nutrient availability, shoot biomass,
and species richness in grassland and wetland communities. Vegetatio 53:121-126.
Walker, J.M. 1959. Vegetation studies on the Delta Marsh, Delta, Manitoba. M.S. Thesis, Univ. of
Manitoba, Winnipeg. 203pp.
Walker, J.M., and R.T. Coupland. 1970. Herbaceous wetland vegetation in the aspen grove and
grassland regions of Saskatchewan. Can. J. Bot. 48:1861-1878.
Walker, J.M., and R.T. Coupland. 1968. An analysis of vegetation-environment relations in
Saskatchewan sloughs. Can. J. Bot. 46:509-522.
Walkley, A. and I.A. Black. 1934. An examination of the Degtjareff method for determining soil organic
matter and a proposed modification of the chromic acid titration method. Soil Sci. 37:251-264.
Ward, P. 1968. Fire in relation to waterfowl habitat of the Delta marshes. Proc. Tall Timbers Fire Ecol.
Conf. 8:254-267.
Ward, E. 1942. Phragmrtes management. Trans. N. Am. Wildl. Conf. 7:294-298.
Wendt, R.C. and R.B. Corey. 1980. Phosphorus variations in surface runoff from agricultural lands as a
function of land use. J. Environ. Qua/. 9:130-136.
Westfall, R.D. 1975. Landowner willingness to allow public access for three recreational activities: a
study of rural landowners in Kent County, Michigan. M. S. Thesis, Michigan State Univ., East
Lansing. 89pp.
Westhoff, V., and E. van der Maarel. 1973. The Braun-Blanquet approach, pp. 617-726 in R. H.
Whittaker (ed.), Ordination and classification of communities. Handbook of Vegetation Science,
Vol. 5. Junk, The Hague.
Whatley, H.E., M.E. Lisano and J.E. Kennamer. 1977. Plasma corticosterone level as an indicator of
stress in the eastern wild turkey. J. Wildl. Manage. 41:189-193.
197
-------
Williams, D.E. 1948. A rapid manometric method for the determination of carbonate in soils. So/7 Sci.
Soc. Proc. 13:127-129.
Wilson, S.D., and P.A. Keddy. 1986. Species competitive ability and position along a natural
stress/disturbance gradient. Ecology 67:1236-1242.
Wingfield, J.C., J.P. Smith and D.S. Farner. 1982. Endocrine responses of white-crowned sparrows to
environmental stress. Condor 84:399-409.
Winter, T.C. and M.R. Carr. 1980. Hydrologic setting of wetlands in the Cottonwood Lake Area,
Stutsman County, North Dakota. Water Res. Invest. 80-89. USGS.
Wolf, A.M., D.E. Baker, H.B. Pionke, and H.M. Kunishi. 1985. Soil tests for estimating labile, soluble,
and algae-available phosphorus in agricultural soils. J. Environ. Qua/. 14:341-348.
Wright, B.A. 1985. An empirical assessment of behavioral aspects and other determinants of rural
landowners' hunter access policies. Ph.D. Diss., Texas A & M Univ., College Station, TX.
186pp.
Zecor, D.T. 1986. The Wisconsin landowner study: landowner attitudes and values toward wildlife
management on private lands, posting, and hunter access. M. S. Thesis, Univ. Wisconsin,
Stevens Point. 54pp.
198
-------
APPENDICES
(Appendices Listed By Chapter; Not All Chapters Have Appendix)
APPENDIX 1-1
Attendance List Interagency Prairie Pothole Workshop July 24-25,1991, Jamestown, ND
Dr. James Arndt, Department of Soil Sciences, North Dakota, State University, P.O. Box 5575, Fargo,
ND 58105
(701) 237-7864
Dr. William T. Barker, Animal Science Department, Hulty Hall, North Dakota State University, Fargo, ND
58105
(701) 237-7222
Mr. Ray Chapman, U.S. Army Corps of Engineers, Waterways, Experiment Station, CEWES-ER-W,
3909 Halls Ferry Road, Vicksburg, MS 39180-6199
(601)634-3774
Mr. Buddy Clairain, U.S. Army Corps of Engineers, Waterways Experiment Station, CEWES-ER-W,
3909 Halls Ferry Road, Vicksburg, MS 39180-6199
(601)634-3774
Dr. Lewis M. Cowardin, Northern Prairie Wildlife Research Center, Rt. 1 Box 26C, Jamestown, ND
58401-9736
(701)252-5363
Mr. Tom Dahl U.S. Fish and Wildlife Service-NWI 9720 Executive Center Drive Monroe Building Suite
#101 St. Petersburg, FL 33702
(813) 893-3624
Dr. Walter Duffey, South Dakota State University, Brookings, SD 57007
(605) 688-6121 (Ext. 4782)
Mr. Mike Ell, North Dakota State Department of Health, Division of Water Quality, P.O. Box 5520,
1200 Missouri Avenue, Bismarck, ND 58502-5520
(701)221-5210
Dr. Ned Euliss, Northern Prairie Science Center, 8711 37th Street Southeast, Jamestown, ND
58401-7317
(701) 252-5363
Mr. Mike Gilbert, COE Regulatory Branch-Omaha District, P.O. Box #5, Omaha, NB 68101
(402) 221-3057
199
-------
Dr. Russ Hall, U.S Fish and Wildlife Service, MS 725 ARLSQ, 1849 C Street, NW, Washington,
DC 20240
(703)358-1710
Mr. Mark Hanson, Wetland Wildlife Population and Research Group, 102 23rd Street, Bemidji,
MN 56601
(218) 755-3920
Dr. Sue Haseltine, Northern Prairie Wildlife Research Center, Rt. 1 Box 26C, Jamestown, ND
58401-9736
(701)252-5363
Mr. William Horak, U.S. Geological Survey, Water Resources Division, 821 E. Interstate Avenue,
Bismarck, ND 58501
(701) 221-5210, FTS 783-4601
Dr. Daniel Hubbard, Dept. of Wildlife and Fisheries, South Dakota State University, P.O. Box
2206, Brookings, SD 57007
(605) 688-6121
Mr. John Jacobson, Ducks Unlimited, 1 Waterfowl Way, Long Grove, IL 60047
(708) 438-4300
Dr. Doug Johnson, Northern Prairie Science Center, 8711 37th Street Southeast, Jamestown, ND
58401-7317
(701) 252-5363; FAX - (701) 252-4217
Mr. Tom Jurik, Department of Botany, Iowa State University, Ames, Iowa 50011
Dr. James LaBaugh, U.S. Geological Survey, Water Resources Division, Mail Stop 413, Building
53, Lakewood, CO 80225
(303) 236-4989; FTS - 776-4989
Ms. Nancy Leibowitz, Mantech Environmental/USEPA, Environmental Research Laboratory, 200
SW 35th Street, Corvallis, OR 97333
(541) 757-4666
Dr. Scott Leibowitz, USEPA Environmental Protection Agency, 200 SW 35th Street, Corvallis, OR
97333
(541) 757-4666
Dr. Richard Nelson, U.S. Bureau of Reclamation, P.O. Box 1017, Bismarck, ND 58502
(701) 250-4593
Mr. Richard Novitzki, Mantech Environmental/USEPA Environmental Research Laboratory, 200
SW 35th Street Corvallis, OR 97333
(541) 757-4666
Mr. John Peters, EPA Region VIII, Water Quality Requirement Section, 99g 18th Street, Suite
500, Denver Place, Denver, CO 80202-2405
FTS - 330-1579
200
-------
Dr. James Richardson, Department of Soil Sciences, North Dakota State University, P.O Box
5575, Fargo, ND 58105
(701) 237-8573
Mr. Donald Rosenberry, U.S. Geological Survey, Water Resources Division, Mail Stop 413,
Building 53, Lakewood, CO 80225
(303) 236-4990; FTS - 776-4990
Ms. Louisa Squires, Mantech Environmental/USEPA Environmental Research Laboratory, 200
SW 35th Street, Corvallis, OR 97333
(541) 757-4666,
Mr. Larry Strong, Northern Prairie Science Center, 8711 37th Street Southeast, Jamestown, ND
58401-9736
(701) 252-5363
Mr. George A. Swanson, Northern Prairie Wildlife Research Center, Rt. 1 Box 26C, Jamestown,
ND 58401-9736,
Dr. Arnold van der Valk, Department of Botany, Iowa State University, Ames, IA 50011
(515) 294-4374,
APPENDIX 1-2
Attendees at Prairie Pothole Interagency Planning Meeting, NPWRC, 9/17/91
Cowardin, Lew
Euliss, Chip
Haseltine, Sue
Johnson, Doug
Leibowrtz, Nancy
Leibowitz, Scott
Medlin, Joel
Novitski, Dick
Preston, Eric
Shaffer, Terry
Swanson, George
Walsh, Dan
NPWRC
NPWRC
NPWRC
NPWRC
MANTECH
EPA
USFWS
MANTECH
EPA
NPWRC
NPWRC
USFWS, Bismarck
201
-------
APPENDIX 2-1
Example of map used for 10.4 km2 Plot 134 in the Prairie Pothole. Numbers are used to identify the
wetland basin. Numbers in parentheses identify the years when ground study was done at these basins.
Source of data was mapping conducted as a special project by the National Wetlands Inventory. The map
data have been corrected for topological errors, and addition of missed wetlands. Errors in wetland
classification have not been corrected because these would require ground survey of all wetland basins.
Linear features have been buffered to average width.
RIGHT OF WAY
TEMPORARY WETLAND
SEASONAL WETLAND
SEMI PERMANENT WETLAND
BARREN
202
-------
APPENDIX 2-2
Sample of SAS output describing individual wet areas derived from the Feature map process.
SAS 9:57 Monday, June 6, 1994 4
BASIN PLOT TYPE IMAGDATE TOTAREA FEATTYPE CENTCOL CENTLINE FEATBOUN _FREQ_ AREA ITEM BASINCLS STRATUM HEALTH
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
3
6
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
NoType
NoType
NoType
HoType
NoType
NoType
HoType
NoType
NoType
HoType
NoType
NoType
PKMA
PEMA
PEMA
PEMA
PEMA
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
FEMC
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
377
139
429
485
474
714
190
620
277
718
49
720
696
519
326
367
94
495
506
625
391
455
471
716
560
522
658
656
624
658
187
143
628
197
516
652
506
382
401
715
197
538
381
47
71
219
248
256
373
384
400
400
407
550
624
176
270
101
643
366
669
511
391
220
608
613
644
461
448
349
366
142
122
332
341
364
658
685
678
332
102
709
187
127
255
277
0.05
0.03
0.02
0.04
0.06
0.10
0.03
0.04
0.03
O.OS
0.07
0.01
0.08
0.08
0.05
0.02
0.07
0.13
0.13
0.05
0.07
0.07
0.04
0.04
0.03
0.01
0.02
0.04
0.08
0.02
0.03
0.02
0.02
0.19
0.01
0.07
0.04
0.27
0.05
0.34
0.08
0.05
0.22
,
,
4
,
.
•
.
.
.
«
.
,
1
1
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
1
3
2
1
1
0.0496
0.0298
0.0198
0.0496
0.0595
0.3274
0.0298
0.0595
0.0298
0.0694
0.1389
0.0099
0.2381
0.1290
1.3194
0.1091
0.1389
0.4662
0.4563
0.0794
0.0992
0.1190
0.0694
0.0595
0.0298
0.0099
0.0099
0.0496
0.2083
0.0099
0.0198
0.0099
0.0198
0.9523
0.0099
0.1290
0.0496
1.9146
0.0694
1.2400
1.9046
0.0694
0.8035
0
0
0
0
0
0
0
0
0
0
0
0
56
98
107
143
511
364
370
372
4
16
17
23
25
27
34
36
46
47
88
89
104
106
120
129
132
146
149
153
159
180
189
.
.
.
.
.
.
.
.
.
.
.
.
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
-------
Appendix 2-2 (Continued)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
PBMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PBMC
PEMC
PEMC
PEMC
PBMC
PEMC
PEMC
PBMC
PEMC
PEMC
FEMC
PEMC
PBMC
PEMC
PEMC
PEMC
FEMC
PEMC
PBMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
FBMC
PEMC
PEMC
PBMC
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
369
388
365
395
404
420
435
366
394
424
371
386
390
421
383
369
365
458
547
459
520
510
539
544
529
460
464
469
478
453
718
713
715
482
461
625
384
409
424
411
510
458
461
504
517
438
424
707
684
691
674
647
632
606
300
311
312
343
336
349
355
334
379
431
421
453
470
509
521
564
661
501
384
372
309
312
159
201
236
216
190
164
158
386
696
335
276
75
452
411
614
97
108
118
705
674
640
648
647
601
575
654
637
691
692
711
714
685
O.OS
0.20
0.04
0.03
0.07
0.11
0.09
0.08
0.70
0.11
0.07
0.06
0.02
0.15
0.54
0.22
0.19
0.13
0.13
0.06
0.16
0.05
0.16
0.08
0.14
0.04
0.02
0.06
0.05
0.07
0.04
0.14
0.05
0.11
0.08
0.08
0.13
0.08
0.11
0.02
0.01
0.10
0.08
0.12
0.13
0.16
0.03
0.13
0.06
0.08
0.04
0.07
0.02
0.05
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
1
1
2
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
0.0992
1.0614
0.0595
0.0298
0.1290
0.2778
0.2381
0.1389
5.2874
0.2579
0.1686
0.0992
0.0198
0.7341
3.0256
0.6746
0.7142
0.4365
0.4365
0.1190
0.4266
0.0794
0.4960
0.2083
0.5158
0.0298
0.0198
0.1190
0.0893
0.0992
0.0397
0.6646
0.0794
5.9718
0.1389
0.1686
0.5456
0.1984
0.2579
0.0198
0.0099
0.3174
0.1488
0.3770
0.3670
0.4762
0.0298
0.3869
0.0992
0.1190
0.0595
0.1389
0.0198
0.0992
190
191
192
194
195
196
197
198
200
201
202
203
204
205
206
207
208
209
211
212
214
215
216
217
218
220
221
222
223
225
236
237
238
239
250
252
275
288
289
290
294
295
296
297
298
299
300
302
303
304
305
306
307
308
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
B
H
H
H
B
B
H
H
B
B
B
B
R
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
-------
Appendix 2-2 (Continued)
ro
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
FEMC
FEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
5S5
601
617
591
592
575
S61
565
551
552
541
531
465
563
523
523
502
679
685
676
692
607
621
591
560
587
584
593
649
655
679
686
625
687
674
656
607
591
586
573
561
554
502
634
701
640
698
682
649
653
673
650
360
209
661
628
621
604
578
563
557
582
588
507
497
512
533
433
460
435
477
376
356
344
310
367
328
374
388
247
260
276
292
322
286
247
175
120
89
112
154
181
197
196
155
242
62
298
280
435
568
560
508
407
395
204
532
651
0.13
0.04
0.08
0.28
0.14
0.16
0.04
0.08
0.10
0.08
0.10
0.16
0.08
0.67
0.09
0.10
0.04
0.10
0.09
0.11
0.26
0.08
0.22
0.27
0.07
0.02
0.05
0.15
0.03
0.13
0.11
0.04
0.22
0.10
0.10
0.09
0.10
0.14
0.13
0.10
0.09
0.08
0.08
0.13
0.03
0.22
0.05
0.07
0.02
0.10
0.13
0.25
0.07
0.09
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.3571
0.0595
0.2083
0.8730
0.5357
0.3274
0.1587
0.1786
0.2877
0.1885
0.2778
0.6250
0.1984
5.7635
0.1190
0.2877
0.2182
0.2976
0.2381
0.3571
1.1408
0.1290
0.4464
1.1606
0.1587
0.0198
0.0992
0.4464
0.0298
0.3869
0.3274
0.0298
1.1706
0.2976
0.2480
0.1786
0.2579
0.4067
0.2976
0.2381
0.1885
0.1786
0.1587
0.2877
0.0298
0.7738
0.1190
0.0992
0.0198
0.1786
0.5357
1.1309
0.1587
0.1488
309
310
311
312
313
314
315
316
317
319
320
321
322
324
325
326
327
328
329
330
331
333
334
335
336
337
338
339
340
341
342
343
344
345
346
348
349
350
351
352
353
354
355
357
358
373
379
380
381
382
383
384
393
398
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
H
H
H
H
B
H
H
H
H
H
H
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
-------
Appendix 2-2 (Continued)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
PEMC
FEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FBMC
FEMC
PEMC
FEMC
FEMC
PEMC
PEMC
FEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
FEMC
PEMC
FEMC
FEMC
PEMC
FEMC
FEMC
FEMC
PEMC
PEMC
PEMC
FEMC
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
59
51
287
114
344
203
189
165
136
129
118
111
101
202
199
204
318
289
282
256
221
238
218
229
247
285
310
289
288
290
258
334
329
353
318
304
277
275
309
322
241
588
321
333
313
265
258
255
354
306
335
306
235
245
462
435
202
208
361
709
54
139
151
181
218
228
251
170
190
214
236
247
243
249
248
263
297
306
302
275
260
312
330
341
341
307
322
338
347
341
361
381
394
379
403
406
417
430
481
451
432
412
518
552
624
652
680
642
0.23
0.13
0.04
0.01
0.31
0.17
0.09
0.07
0.01
0.03
0.03
0.07
0.16
0.01
0.01
0.07
0.07
0.02
0.02
0.32
0.08
0.08
0.04
0.10
0.01
0.20
0.05
0.20
0.07
0.05
0.16
0.07
0.11
0.05
0.04
0.05
0.02
0.10
0.07
0.05
0.16
0.04
0.09
0.06
0.13
0.11
0.07
0.08
0.10
0.53
0.06
0.10
0.04
0.08
1
1
1
1
1
1
1
1
3
1
2
1
1
2
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
2
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1.4483
0.4662
0.0595
0.0099
0.8134
0.6944
0.1984
0.0992
0.6746
0.0298
0.0397
0.1686
0.5754
0.0198
0.0099
0.1290
0.1587
0.0198
0.0198
1.6765
0.2182
0.1984
0.0298
0.2976
0.0198
0.8730
0.0893
0.8134
0.1190
0.0794
0.6845
0.7638
0.3075
0.0595
0.0397
0.0694
0.0099
0.3075
0.1786
0.1587
0.2877
0.0496
0.2480
0.1091
0.4067
0.3373
0.1190
0.1885
0.1686
3.6406
0.0992
0.1290
0.0496
0.2381
406
407
408
413
415
419
420
425
426
427
428
429
430
432
433
434
435
436
437
438
440
441
443
444
445
446
447
448
449
450
452
453
455
456
457
458
459
461
462
463
465
466
467
468
470
471
472
473
474
475
477
478
479
480
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
R
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
B
H
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
-------
Appendix 2-2 (Continued)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
8
9
10
17
26
28
112
116
0
0
0
0
0
0
0
0
0
0
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
PEMC
PEMC
PBMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
FEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PEMC
PABF
PEMF
PUBFX
PEMF
PUBFX
PUBFX
PABF
PEMF
PEMC
PEMF
PEMF
PEMF
PEMF
PEMF
PEMF
PEMF
PEMF
PEMF
PEMF
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wat
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wat
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
158
130
185
130
104
63
77
101
110
122
137
162
192
189
164
128
166
148
122
206
168
166
151
171
147
197
179
258
228
269
139
286
261
328
266
518
59
276
146
364
543
698
178
358
692
614
395
489
499
491
550
573
594
639
614
538
524
501
505
409
442
450
461
433
455
447
448
412
395
389
374
371
340
319
331
292
274
251
251
676
689
77
102
198
48
51
48
707
714
388
49
147
702
230
360
145
641
688
226
70
589
711
575
630
659
385
215
96
0.13
0.08
0.12
0.18
0.03
0.04
0.02
0.10
0.03
0.11
0.17
0.16
0.13
0.23
0.12
0.08
0.08
0.40
0.23
0.12
0.10
0.16
0.16
0.04
0.09
0.08
0.02
0.08
0.04
0.05
0.06
0.14
0.05
0.07
0.05
0.02
0.08
0.19
0.07
0.10
0.05
0.55
0.37
0.06
0.15
0.47
0.40
0.19
0.42
0.19
0.58
0.10
0.52
0.16
1
1
1
1
1
2
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
1
1
1
1
1
1
1
3
2
1
2
2
1
1
6
11
3
4
1
1
1
1
1
1
1
1
0.2678
0.1686
0.4365
0.7539
0.0298
0.0694
0.0099
0.2182
0.0298
0.4266
0.6150
0.7043
0.5555
1.2797
0.4662
0.1786
0.1984
1.8451
0.9226
0.3869
0.3075
0.4762
0.6944
0.0397
0.1488
0.1389
0.0198
0.3670
0.0397
0.0794
0.1190
0.3968
0.0992
0.1389
0.0694
11.3187
3.0157
0.4464
12.2214
0.3373
0.0694
4.8112
12.1718
2.2518
4.5334
3.2736
2.0038
0.8333
3.7597
0.9226
4.3549
0.1984
4.3549
0.8134
481
483
484
486
487
488
489
491
492
493
494
495
496
497
498
499
500
501
502
504
505
506
508
509
510
517
518
528
529
530
554
555
556
575
576
241
549
526
105
84
183
367
392
395
152
162
279
280
282
283
284
285
286
287
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
B
B
B
B
B
B
H
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
H
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
-------
Appendix 2-2 (Continued)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
PABF
PABF
PABF
PEMF
PEMF
PEMF
PABF
PEMF
PBMF
PEMF
PEMF
PEMF
PUBFX
PABF
PABF
PUBFX
PDBFX
PUBFX
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
12174
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
2570.61
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
wet
430
619
425
676
667
693
662
55
356
58
94
194
132
292
231
560
478
272
705
119
660
594
549
516
165
498
256
237
394
258
309
654
464
357
192
335
0.25
0.33
0.35
0.87
0.97
0.21
0.32
0.31
0.10
0.26
0.81
0.13
0.08
0.34
0.32
0.08
0.09
0.12
1
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
1.5178
2.3312
3.2339
8.6205
6.3091
0.4960
2.4205
2.2518
0.2480
1.4880
6.2000
0.4067
0.1290
1.6666
2.7478
0.2083
0.1686
0.4067
366
368
369
374
375
376
385
404
405
410
417
418
512
513
514
577
578
579
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
H
B
H
H
B
B
B
B
B
B
B
B
B
B
B
B
B
B
"BASIN = Basin number from vector, PLOT = EMAP plot number, TYPE = Cowardin wetland class, IMAGDATE = Date of videography in SAS format,
TOTAREA = Total area of plot in acres, FEATTYPE = Generic feature type from feature map, CENTCOL = Screen column location of centroid of
feature, CENTLINE = Screen line location of centroid of feature, FEATBOUN = Summary boundary length of feature in miles, _FREQ_ = Number of
ponds in a basin, AREA = Total area of water in a basin, ITEM = MIPS internal polygon id number, BASINCLS = Cowardin numerical basin class,
STRATUM = Stratum of EMAP plot based on original draw, HEALTH = health of EMAP plot based on original draw.
-------
APPENDIX 4-1
MIPS PROCEDURES FOR NWI DIGITAL DATA
IMPORT PROCEDURE
Basic procedures for importation and editing of NWI MOSS data for EMAP are described
below. Data for the original plots came as a three part digital data set in MOSS format with UTM
projection for each of the 4 square mile plots. The data sets are polygon data, buffer data (line and
point data buffered by NWI in MOSS), and line data (road linears). All data are imported into MIPS
with redundant lines removed out to a distance of 1.1 vector units.
LINE BUFFER DISTANCES AND PROCEDURES
Gaps in road linears and additions or deletions of roads were determined from
photointerpretation and ground truth information. Line data was then hand edited to remove these gaps
or add new information. Line data for roads were double buffered for both the road surface and the
right of way. The polygon, buffer and buffered road data sets were then intersected.
BUFFER ZONE DISTANCES
This procnote is extracted from PROCNOTE.103.
(8) Riqht-of-way.--The area between road surface and the fence in grassland and between road
surface and cropland in farmed areas. The cover of road right-of-ways is variable but often consists of
smooth brome (Bromus inermis), a cool season grass that will show active growth at the time of
photography. Only very large right-of-ways such as some interstate highway and some railroads will be
large enough to delineate as a polygon. Narrower roads should be mapped as labeled linear features.
These linear features will be converted to areas during digitization so class 8 (right-of-way) must be
subdivided into subclasses as follows:
8a gravel road 8e dirt road
8b hard surface road 8d fencerow
8c railroad
209
-------
Prairie trails appearing as tire tracks across grassland areas should be ignored, not mapped as
right-of-way.
8a gravel road
right-of-way
road surface
right-of-way
right-of-way
road surface
right-of-way
right-of-way
road surface
right-of-way
right-of-way
13'
20'-
I
13'
I
62.5'
I
I
25'--
I
I
62.5'
I
65'
I
20'-
I
65'
I
8A
g
9
8A
8b hard surface road
8c railroad
8d fence rows and field borders
10'-
210
-------
8E dirt road-no right of way
road surface only
16'
HAND EDIT OF INTERSECTED DATA SET
Hand editing of the combined vector set was done to fix overlapping areas of differing polygon
classes and final check and edit of basin numbers.
CRITERIA FOR OVERLAPS
1) Assume that center of road or edge of leld is where study plot begins and ends. Delete lines
outside the center line OR the road if that line contains the ownership classification.
2) When buffering overlays, the polygons with the original data would take precedence, eg.,
buffered line overlays original polygon. Remove the buffered line where it intersects the original
polygon.
3) Right-of-way takes precedence over all upland classifications.
4) Wetland will take precedence over everything.
5) If wetland overlays wetland than the most permanent water regime takes precedence.
6) Buffered road surfaces (9) will take precedence over wetlands.
7) When a river crosses a buffered road take out the road right-of-way, but leave the road. The
river will go up to the road poly and continue on.
211
-------
RULES FOR DEFINING WETLAND BASINS
These rules are taken from my notes from a meeting between Cowardin, Pywell, and Gebhard
at St. Petersburg September 11,1986.
If two basins are connected by a linear wetland each basin and the connecting link are treated
as separate basins if the water regime of the connecting link is temporary (a). Other wise the basins
and the connecting link are all considered one basin. If two basins are connected by a ditch (x
modifier) the ditch and the two ends are considered separate basins regardless of the water regime of
the ditch. If there is an obvious difference between the elevation of the basins when viewed in stereo
the basins and any connecting link are considered separate basins.
When working with the data sets for the plots we noted some errors in the interpretation of
basins. In some cases a temporarily flooded arm extending from the edge of a basin wetland was
treated as a separate basin. These arms should have been included in the basin. In some cases
where basin numbers were being added to polygons, the same basin number was assigned to
polygons in two obviously separate basins.
Stream orders: one basin unless its broken by a road or the stream order changes. Ex. a
small stream enters a larger stream. It would not be the same basin, there would now be three basins.
Any wetland that is continuous within a riverine system would be considered the same as the
riverine basin.
FINAL FIXES
The final intersect done with the vector data includes the updates for wetlands missed by NWI
or not present on the original photography (See EMAP 004.DOC).
The final edit of the vector data included changes in upland cover for Conservation Reserve
Program (CRP). CRP information was obtained from county ASCS offices and edited in the final vector
set. The final vector set for each plot was placed in EMAPPOLY.RVF.
212
-------
APPENDIX 4-2
VIDEO CAPTURE
Video tape information was directly captured into digital form on personal computers. Since the
video was originally taken with a black and white camera with a near infrared filter, data was captured
as an 8 bit grayscale image.
IMAGE RECTIFICATION AND MOSAIC OF VIDEO IMAGES
a) For each section, georeference the image with NWI vector data using calibrate raster with
vector process. First use linear least squares fit to get a near fit to the image. Second, go to a
piecewise linear model to add and delete control points to get a final fit of the vectors to the
image.
b) Save the control point list and raster cell size upon exiting.
c) Warp and resample the image using the piecewise plane projective model and the control point
list. Use the default value of 20 grids and specify a cell size approximately equal to the input
raster cell size. If the resampled image is still distorted, go back to the original raster and try a
2nd order polynomial fit and edit control points. Warp and resample the image to a 2nd order
polynomial fit.
d) Mosaic the sections using the georeference information to create a single video image for each
plot for each month.
e) If the image is in pieces too small to obtain sufficient control points, put together a large enough
image using the manual positioning process of mosaic and the go to step a).
FEATURE MAP PROCESSING OF VIDEO IMAGES
Feature map processing used in MIPS is a semiautomated, on screen interpretive method of
delineating "features." For more information, see "Feature Mapping Application Note for the Map and
Image Processing System." Delineation of water availability was the primary task of feature mapping in
213
-------
EMAP. Products from the feature map process include a saved screen of the classified image, an 8 bit
image consisting of only the classified features and a text file with areas, boundary lengths, NWI
polygon classes, and NWI basin numbers of each feature.
The actual analysis of each scene combined both automated processing and photointerpretation.
Water has great absorbance in the near infrared range and appeared as black or almost black in the
scene. A scene was displayed on screen then a point or range test was conducted using apparent
water as the sample pixels. A point test highlights every pixel in the scene that has the same color
number as the sample pixel. The range test highlights every pixel in the scene that has a color number
that falls in the range between the highest color number and lower color number of all the sample
pixels. After some number of sample selections and iterations, all readily apparent water areas had
been highlighted. Further manual definition of water boundaries was then done using the drawing
tools. Ground truth information from the field teams was used as additional information for during
mapping. Areas of water underneath dense vegetation, smaller dugouts and stock ponds and riverine
areas could be mapped in this manner.
After all features were classified and delineated, the feature map translabel process was run.
This process involves "rubber sheeting" the NWI vector data over the classified scene. Every feature
was then matched with the corresponding NWI polygon and NWI wetland class and basin number was
transferred to the output file. If a feature was not matched to an NWI polygon, the basin number is 0
and the class is NoType in the output file.
The feature map output raster was converted to a vector object and information about the UTM
centroid of the feature was output to an ASCII file.
FILE NAMING CONVENTIONS
ORIGINAL videography is:
pppSSmmy.RVF
ppp = PLOT NUMBER
SS = WHICH SECTION OF THE PLOT (E.g. NW for Northwest)
mm =MONTH
y OR yy = YEAR
214
-------
The convention for warped individual scenes is
PpppWARP.rvf
The convention for mosaicked images for the entire plot is:
Mpppmmyy.RVF
The convention for feature mapped saved screen images of the entire plot is:
Fpppmmyy.RVF
The convention for feature map output rasters from the entire plot is:
pppmmyyF.RVF
The convention for text file output from feature map is:
Dpppmmyy.txt
The convention for centroid text data from vectorized feature map rasters is
CFpppmyy.txt
215
-------
APPENDIX 4-3
IMAGE RECTIFICATION AND MOSAIC OF PHOTO IMAGES
a) Two slides provided coverage of one section from high level photography. MIPS manual
mosaic process was used to create one raster for each section.
b) For each section, georeference the image with NWI vector data using calibrate raster with
vector process. First use linear least squares fit to get a near fit to the image. Second, go to a
piecewise linear model to add and delete control points to get a final fit of the vectors to the
image.
c) Save the control point list and raster cell size upon exiting.
d) Warp and resample the image using the piecewise plane projective model and the control point
list. Use the default value of 20 grids and specify a cell size approximately equal to the input
raster cell size. If the resampled image is still distorted, go back to the original raster and try a
2nd order polynomial fit and edit control points. Warp and resample the image to a 2nd order
polynomial fit.
e) Mosaic the sections using the georeference information to create a single photo image for each
plot.
The convention for photography information for the high altitude scenes is:
Spppss92.rvf
Where ppp is the plot number and ss is the slide number.
See Procnotes 920805tb and 930804tb for more information.
The convention for warped, mosaicked full plot photos is:
Wpppmmyy.rvf
Where ppp is the plot number mm is the month of the photography and yy is the year.
216
-------
APPENDIX 4-4
UPDATE OF NW1 VECTORS
It was apparent from 1992 video and photography that there were some errors associated with
NWI vector data. To update these vectors and also provide some measure of probable drainage within
the EMAP study plots, a review and photointerpretation of all the video images and the photography
was attempted. Based on this review, there were five classes of polygons and two line classes
created. Vector polygons were created by editing a new vector set over a raster with embedded
existing NWI wetland vectors.
For polygon data that was not on the NWI data sets:
1) If an area held water for 2 or more months and did not have an obvious berm, the
class was "PEMC."
2) If an area held water for 2 or more months and did have a visible berm, the class was
"PUBFX."
3) If an area held water for 2 or more months and had an obvious dam structure in a
watershed, the class was the class of the riverine system with an added modifier "H"
(e.g. if stream was R2UBG then stock pond was R2UBGh)
4) If an area was wet for at least one month, had an obvious basin shape in the other
months when it was dry and had an obvious drainage channel from it, then the class
was "PEMAd."
5) If an area kept water all four months and did not show any visible berm, the class was
"PEMF."
217
-------
For linear data not on the NWI data sets, another vector object of lines was created, then
buffered out 24 feet.
1) If there is an obvious artificial or enhanced natural drainage between two or more
wetlands and is visible as water at least one month and a clear line for at least two
months, then the line class is "drainage". Drainage and ditch areas were analyzed
using all available data. If a new ditch in a plot appeared to have all the characteristics
of other NWI mapped ditches (vegetation, water, etc.), the ditch was considered a ditch
rather that casual drainage or a grassed waterway. The line representing the ditch was
buffered out to a radius of 24 feet, a class "PEMCx" was given to the ditch and it was
added to the NWI vectors.
2) If there is an obvious natural drainage that is not on NWI data then the line class is
"natural drainage."
GRASSED WATERWAYS BUFFER DISTANCE
Certain plots in southern part of the prairie pothole region have drainage systems from wetland
areas that cannot be classified as wetland. They consist primarily of grassed waterways for runoff. In
order to account for the area of this grass within a plot, an estimate was made of the width of these
grass areas from both aerial photography and where NWI had digitized some of these areas. It was
estimated to be a radius of 15 feet on each side of the centerline of a grassed waterway. A vector line
file was created that reflected the grassed waterways and then buffered out to a radius of 15 feet. This
grassed area was given a Cowardin upland class 7.
FINAL INTERSECT
The new polygons, buffered linear wetlands and grassed waterways created in this procedure were
intersected and edited according to criteria established in EMAP001 .DOC for overlaps in vector data.
This data set was named UPDATES.RVF. and used in the final intersect with EMAPPOLY.RVF.
218
-------
APPENDIX 4-5
VEGETATION ZONE DELINEATION
The following procedures were used to create vector data from vegetation zones determined by
Hal Kantrud (NPSC).
1) Each low level slide was printed to Cibachrome paper.
2) Hal Kantrud annotated the photo in the field in his survey of EMAP wetland with
vegetation zones.
3) Each photo was scanned into MIPS RVF format.
4) Each raster was georeferenced with high elevation slide mosaics (EMAP003.DOC)
using raster to raster registration using ground control points in common on both
rasters.
5) Vectors were edited over the georeferenced photo rasters to delineate vegetation zones
and place a point where each vegetation quadrat and soil sample were taken.
6) Information on the area and boundary of each zone was then exported to an ASCII text
file for further processing.
APPENDIX 4-6
WATERSHED UPLAND DELINEATION
Watershed delineation was completed for the EMAP sample wetlands. The soils and
vegetation team used a compass and hand held clinometer at varying stations within the wet meadow
zone. From each station, the bearing from north and elevation angle to the highest point on the
horizon was recorded and the distance to that point was paced by the scientist. Using trigonometry
219
-------
and the elevation angle, the horizontal distance was calculated from each station to the apparent
horizon. Within MIPS, a point for each station was drawn on the georeferenced raster of the basin and
the surrounding area from low level photographs annotated by the vegetation and soils team. From
that point, both distance and angle from north were plotted to an outside point for each corrected
measurement. A georeferenced watershed vector was then created by connecting each of the outside
points while following visible contours and/or information for a contour map of the same area. Once a
watershed outline was created, a evaluation of uplands adjacent to the wetland was done by the
principal investigators of landscape ecology and vegetation and the vectors edited to reflect the upland
information. Classes followed Cowardin et.al. 1979. An vector intersection of the watershed uplands
was done with the wetland vector information provided by the vegetation biologist. Areas of the
watershed uplands and linear distance adjacent to the wetland boundary were computed and exported
to SAS.
220
-------
Appendix 6-1
Plants3 recorded in sampled communities in EMAP wetlands, 1992-1993.
Svmbolb
*
*
#
#
*
*
*
*
*
*
!
*
!
tt
*
#
*
j
!
*
tt
*
tt
*
*
•*
*
tt
*
*
*
!
#
*
#
#
*
*
*
*
Code
000000
ACNB2
ACMI2
AGCA2
AGCR
AGRE2
AGSM
AGHY
AGSC5
ALG
ALGR
ALPL
ALCA3
ALAE
AMRE
AMAR2
AMPS
AMFR
ANGE
ANCA8
AN
APS I
AR
ARAB3
ARBI2
AEFR4
ARLU
AS
ASC
AS IN
ASOV
ASSP
ASSY
ASBR3
ASER3
ASSI2
ATPA4
AVSA
BESY
BIFR
BOAS
BR
BRIN2
BRJA
CAAR18
CACA19
CACA4
CAIN
CAVE2
Life
Scientific name Prairie
NO VASCULAR PLANTS
ACER NEGUNDO
ACHILLEA MILLEFOLIUM
AGROPYRON CANINUM
ACROPYRON CRISTATUM
AGROPYRON RSPENS
AGROPYRON SMITHII
AGROSTIS HYEMALIS
AGROSTIS SCABRA
MACROALGAE UNIDENTIFIED
ALISMA GRAMINEUM
ALISMA PLANTAGO-AQUATTCA
ALLIUH CANADENSE
ALOPECURUS AEQUALIS
AMARANTHUS RETROFLEXUS
AMBROSIA ARTEMISIIFOLIA
AMBROSIA PSILOSTACHYA
AMORPHA FRUTICOSA
ANDROPOGON GERARDII
ANEMONE CANADENSIS
ANTENNARIA SP. UNIDENTIFIED
APOCYNUM SIBIRICUM
ARABIS UNIDENTIFIED
ARTEMISIA ABSINTHIUM
ARTEMISIA BIENNIS
ARTEMISIA FRIGIDA
ARTEMISIA LUDOVICIANA
ASTER SP UNIDENTIFIED
ASCLEPIAS UNIDENTIFIED
ASCLEPIAS INCARNATA
ASCLEPIAS OVALIFOLIA
ASCLEPIAS SPECIOSA
ASCLEPIAS SYRIACA
ASTER BRACHYACTIS
ASTER ERICOIDES
ASTER SIMPLEX
ATRIPLEX PATULA
AVENA SATIVA
BECKMANNIA SYZIGACHNE
BIDENS FRONDOSA
BOLTONIA ASTEROIDES
BRASSICACEAE UNIDENTIFIED
BROMUS INERMIS
BROMUS JAPONICUS
CARAGANA ARBORESCENS
CARUM CARVI
CALAMAGROSTIS CANADENSIS
CALAMAGROSTIS INEXPANSA
CALLITRICHE VERNA
history in
Pothole reqion0
NP
NP
NP
IP
IP
NP
NP
NP
-
NP
NP
NP
NP
NA
NA
NP
NP
NP
NP
-
NP
-
IP
IA
NP
NP
-
-
NP
NP
NP
NP
NA
NP
NP
NA
IA
NA
NA
NP
-
IP
IA
IP
NP
NP
NP
NP
221
-------
Appendix 6-1 (continued)
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
#
*
*
*
tt
ft
*
*
*
tt
!
*
#
#
*
*
*
*
*
*
*
ft
*
*
*
*
*
*
*
tt
*
ft
*
CAPA5
CAMI2
CABU2
CA
CAAQ
CAAT3
CAAT2
CABE2
CABR10
CAGR3
CAHA3
CAIN11
CALA12
CALA30
CAPR5
CARO6
CARE4
CASA8
CASC11
CAVU2
CEDE4
CHAR
CH
CHAL7
CHRU
CIMA2
CIAR4
CIFL
CIVIT
COAR4
COCAS
COST4
CRRU3
CYOF
CY
DECA5
DERI 2
DES02
DISP
DR
ECCR
ELAC
ELC02
ELEN
ELER
ELMA5
ELSM
ELXY
ELCA4
EPLE2
EQAR
EQHY
EQLA
ERPH
ERP06
ERGA
EUMA12
EUES
EUMA7
CALTHA PALUSTRIS NP
CAMELINA MICROCARPA IA
CAPSELLA BURSA-PASTORIS NA
CAREX UNIDENTIFIED
CAREX AQUATILIS NP
CAREX ATHROSTACHYA NP
CAREX ATHERODES NP
CAREX BEBBII NP
CAREX BREVIOR NP
CAREX GRANULARIS NP
CAREX HALLII NP
CAREX INTERIOR NP
CAREX LAEVICONICA NP
CAREX LANUGINOSA NP
CAREX PRAEGRACILIS NP
CAREX ROSTRATA NP
CAREX RETRORSA NP
CAREX SARTWELLII NP
CAREX SCOPARIA NP
CAREX VULPINOIDEA NP
CERATOPHYLLUM DEMERSUM NA
CHARA SPP.
CHENOPODIUM SP.
CHENOPODIUM ALBUM IA
CHENOPODIUM RUBRUM NA
CICUTA MACULATA NP
CIRSIUM ARVENSE IP
CIRSIUM FLODMANII NP
CIRSIUM VULGARE IP
CONVOLVULUS ARVENSIS IP
CONYZA CANADENSIS NA
CORNUS STOLONIFERA NP
CREPIS RUNCINATA NP
CYNOGLOSSUM OFFICINALE IP
CYPERUS SP. UNIDENTIFIED
DESCHAMPSIA CESPITOSA NP
DESCURAINIA RICHARDSONII NA
DESCURAINIA SOPHIA IA
DISTICHLIS SPICATA NP
DREPANOCLADUS SP.
ECHINOCHLOA CRUSGALLI IA
ELEOCHARIS ACICULARIS NP
ELEOCHARIS COMPRESSA NP
ELEOCHARIS ENGELMANNII NA
ELEOCHARIS ERYTHROPODA NP
ELEOCHARIS MACROSTACHYA NP
ELEOCHARIS SMALLII NP
ELEOCHARIS XYRIDIFORMIS NP
ELYMUS CANADENSIS NP
EPILOBIUM LEPTOPHYLLUM NP
EQUISETUM ARVENSE NP
EQUISETUM HYEMALE NP
EQUISETUM LAEVIGATUM NP
ERIGERON PHILADELPHICUS NP
ERIOPHORUM POLYSTACHION NP
ERUCASTRUM GALLICUM IA
EUPATORIADELPHUS MACULATUS NP
EUPHORBIA ESULA IP
EUPHORBIA MACULATA NA
222
-------
Appendix 6-1 (continued)
*
*
ft
*
!
*
ft
ft
*
*
*
*
#
ft
ft
*
*
ft
*
ft
*
*
*
!
*
*
ft
*
*
*
*
*
*
*
ft
*
*
*
*
*
ft
EUGR5
F01
F02
F03
F04
FOB
FRVI
FRPE
GA
GABO2
GATR2
GEAL3
GLMA3
GLST
GLMA4
GLLE3
GR
GRNE
GRSQ
HADE
HEAU
HEAN3
HEMA2
HENU
HERI2
HEHE5
HEMA3
HIVU2
HOJU
HOVU
HYMA2
HYHI2
IMCA
IR
IVXA
JU
JUBA
JUTED
JUIN2
JUTO
KOSC
LA
LAPU
LASE
LAEC
LAPA4
LEMI3
LETR
LEDE
LIPY
LIAQ
LOSP
LOPU3
LYAM
LYAS
LYCI
LYHY
LYTH2
MARO
EUTHAMIA GRAMINIFOLIA NP
FORB UNIDENTIFIED NO. 1
FORB UNIDENTIFIED NO. 2
FORB UNIDENTIFIED NO. 3
FORB UNIDENTIFIED NO. 4
FORB UNIDENTIFIED NO. 5
FRAGARIA VIRGINIANA NP
FRAXINUS PENNSYLVANIA NP
GALIUM UNIDENTIFIED
GALIUM BOREALE NP
GALIUM TRIFIDUM NP
GEUM ALEPPICUM NP
GLYCERIA MAXIMA NP
GLYCERIA STRIATA NP
GLYCINE MAX IA
GLYCYRRHIZA LEPIDOTA NP
GRAMINEAE UNIDENTIFIED
GRATIOLA NEGLECTA NA
GRINDELIA SQUARROSA NP
HACKELIA DEFLEXA NP
HELENIUM AUTUMNALE NP
HELIANTHUS ANNUUS NA
HELIANTHUS MAXIMILIANI NP
HELIANTHUS NUTTALLII NP
HELIANTHUS RIGIDUS NP
HELIOPSIS HELIANTHOIDES NP
HESPERIS MATRONALIS IA
HIPPURIS VULGARIS NP
HORDEUM JUBATUM NP
HORDEUM VULGARE IA
HYPERICUM MAJUS NP
HYPOXIS HIRSUTA NP
IMPATIENS CAPENSIS NA
IRIS SP.
IVA XANTHIFOLIA NA
JUNCUS SP.
JUNCUS BALTICUS NP
JUNCUS TENUIS NP
JUNCUS INTERIOR NP
JUNCUS TORREYI NP
KOCHIA SCOPARIA IA
LABIATAE UNIDENTIFIED
LACTUCA PULCHELLA NP
LACTUCA SERRIOLA IA
LAPPULA ECHINATA IA
LATHYRUS PALUSTRIS NP
LEMNA MINOR
LEMNA TRISULCA
LEPIDIUM DENSIFLORUM NA
LIATRIS PYCNOSTACHYA NP
LIMOSELLA AQUATICA NP
LOBELIA SPICATA NP
LOTUS PURSHIANUS NP
LYCOPUS AMERICANUS NP
LYCOPUS ASPER NP
LYSIMACHIA CILIATA NP
LYSIMACHIA HYBRIDA NP
LYSIMACHIA THYRSIFLORA NP
MALVA ROTUNDIFOLIA IA
223
-------
Appendix 6-1 (continued)
*
#
*
*
*
*
*
*
*
*
*
#
*
*
*
*
*
*
*
*
*
ft
*
*
*
*
*
*
*
*
*
*
*
*
tt
*
*
*
#
*
*
*
*
*
j
i
*
*
*
MAVE2
MELU
MESA
MB
MEOF
MEAR4
MO
MURI
MYMI2
MYSP2
NECA2
OEBI
ORLU2
OXST
PAVI2
PAVI5
PECA
PESA5
PHAR3
PHPR3
PHVI5
PIPU2
PLSC2
PLMA2
POCO
POPA2
POPR
POAM8
POAV
POC010
POER2
POHY
POLA4
POPE2
PORA3
POBA2
PODE3
POTR10
PO6R8
POPE6
POAN5
POAR7
PONO3
POPES
PORI3
PUNU2
PYVI
RACO3
RAFL
RAGM
RAMA2
RASC3
RATR
RIFL
RINA
RIAM2
ROPA2
ROARS
ROBL
MARSILEA VESTITA NP
MEDICAGO LUPULINA IA
MEDICAGO SATIVA IP
MELILOTUS UNIDENTIFIED
MELILOTUS OFFICINALIS IA
MENTHA ARVENSIS NP
MOSS, UNIDENTIFIED
MUHLENBERGIA RICHARDSONIS NP
MYOSURUS MINIMUS NA
MYRIOPHYLLUM SPICATUM NP
NEPETA CATARIA IP
OENOTHERA BIENNIS NA
ORTHOCARPUS LUTEUS NA
OXALIS STRICTA NP
PANICUM VIRGATUM NP
PARTHENOCISSUS VITACEA NP
PEDICULARIS CANADENSIS NP
PETASITES SAGITTATUS NP
PHALARIS ARUNDINACEA NP
PHLEUM PRATENSE IP
PHYSALIS VIRGINIANA NP
PILEA PUMILA NA
PLAGIOBOTHRYS SCOULERI NA
PLANTAGO MAJOR NP
POA COMPRESSA IP
POA PALUSTRIS NP
POA PRATENSIS IP
POLYGONUM AMPHIBIUM NP
POLYGONUM AVICULARE NA
POLYGONUM CONVOLVULUS NA
POLYGONUM ERECTUM NA
POLYGONUM HYDROPIPER IA
POLYGONUM LAPATHIFOLIUM NA
POLYGONUM PENSYLVANICUM NA
POLYGONUM RAMOSISSIMUM NA
POPULUS BALSAMIFERA NP
POPULUS DELTOIDES NP
POPULUS TREMULA NP
POTAMOGETON GRAMINEUS NP
POTAMOGETON PECTINATUS NP
POTENTILLA ANSERINA NP
POTENTILLA ARGUTA NP
POTENTILLA NORVEGICA NA
POTENTILLA PENSYLVANICA NP
POTENTILLA RIVALIS NA
PUCCINELLIA NUTTALLIANA NP
PYCNANTHEMUM VIRGINIANUM NP
RATIBIDA COLUMNIFERA NP
RANUNCULUS FLABELLARIS NP
RANUNCULUS GMELINII NP
RANUNCULUS MACOUNII NA
RANUNCULUS SCELERATUS IA
RANUNCULUS TRICHOPHYLLUS NP
RICCIA FLUITANS
RICCIOCARPUS NATANS
RIBES AMERICANUM NP
RORIPPA PALUSTRIS NA
ROSA ARKANSANA NP
ROSA BLANDA NP
224
-------
Appendix 6-1 (continued)
*
*
*
*
*
*
*
!
*
*
*
*
*
*
*
*
*
*
*
#
*
*
i
*
i
#
tt
*
*
*
ft
*
*
*
*
*
*
*
*
#
*
*
*
*
*
*
#
*
*
»
*
ft
*
ROWO
RU
RUHI2
RUCR
RUMA4
RUME2
RUOC3
RUOR2
RUST4
SA
SACU
SALA2
SARU
SAAM2
SAEX
SC
SCGA
SCAC
SCAT2
SCFL
SCHE
SCMA
SCPU3
SCVA
SCFE
8E
SECO2
SI
SIAR4
SILOS
SISU2
SOCA6
SOGI
SOMI2
SORI2
SOPT3
SORO
SOAR2
SPEU
SPPE
SPOB
SPPO
STPA
SUDE
SYOC
TAOF
TECA3
THDA
THAR5
TORY
TRBR
TRDU
TRRE3
TRMA4
TRAE
TYAN
TYGL
TYLA
ULPU
ROSA WOODSII up
RUMEX UNIDENTIFIED
RUDBECKIA HIRTA NP
RUMEX CRISPUS ip
RUMEX MARITIMUS NA
RUMEX MEXICANUS NP
RUMEX OCCIDENTALIS NP
RUMEX ORBICULATUS NP
RUMEX STENOPHYLLUS IP
SALIX UNIDENTIFIED
SAGITTARIA CUNEATA NP
SAGITTARIA LATIFOLIA NP
SALICORNIA RUBRA NA
SALIX AMYGDALOIDES NP
SALIX EXIGUA NP
SCUTELLARIA UNIDENTIFIED
SCUTELLARIA GALERICULATA NP
SCIRPUS ACUTUS NP
SCIRPUS ATROVIRENS NP
SCIRPUS FLUVIATILIS NP
SCIRPUS HETEROCHAETUS NP
SCIRPUS MARITIMUS NP
SCIRPUS PUNGENS NP
SCIRPUS VALIDUS NP
SCOLOCHLOA FESTUCACEA NP
SETARIA SP. UNIDENTIFIED
SENECIO CONGESTUS NA
SILENE UNIDENTIFIED
SINAPSIS ARVENSIS IA
SISYMBRIUM LOESELII NA
SIUM SUAVE NP
SOLIDAGO CANADENSIS NP
SOLIDAGO GIGANTEA NP
SOLIDAGO MISSOURIENSIS NP
SOLIDAGO RIGIDA NP
SOLANUM PTYCANTHUM NA
SOLANUM ROSTRATUM NA
SONCHUS ARVENSIS IP
SPARGANIUM EURYCARPUM NP
SPARTINA PECTINATA NP
SPHENOPHOLIS OBTUSATA NP
SPIRODELA POLYRHIZA
STACHYS PALUSTRIS NP
SUAEDA DEPRESSA NA
SYMPHORICARPOS OCCIDENTALIS NP
TARAXACUM OFFICINALE IP
TEUCRIUM CANADENSE NP
THALICTRUM DASYCARPUM NP
THLASPI ARVENSE IA
TOXICODENDRON RYDBERGII NP
TRADESCANTIA BRACTEATA NP
TRAGOPOGON DUBIUS IP
TRIFOLIUM REPENS IP
TRIGLOCHIN MARITIMUM NP
TRITICUM X AESTIVUM IA
TYPHA ANGUSTIFOLIA IP
TYPHA X GLAUCA
TYPHA LATIFOLIA NP
ULMUS PUMILA IP
225
-------
Appendix 6-1 (continued)
*
*
*
*
*
!
*
*
#
*
#
*
*
URDI
UTMA
VEBR
VEHA2
VEFA2
VEPE2
VI
VIAM
VIRI
XAST
ZAPA
ZEMA
ZIEL2
ZIAP
URTICA DIOICA
UTRICULARIA MACRORHIZA
VERBENA BRACTEATA
VERBENA HASTATA
VERNONIA FASCICULATA
VERONICA PEREGRINA
VICIA UNIDENTIFIED
VIC I A AMERICANA
VITIS RIPARIA
XANTHIUM STRUMARIUM
ZANNICHELLIA PALUSTRIS
ZEA MAYS
ZIGADENUS ELE6ANS
ZIZIA APTERA
NP
NA
NA
NP
NP
NA
-
NP
NP
NA
-
IA
NP
NP
"Nomenclature follows Great Plains Flora Association (1992), so some listed taxa are synonyms in United States Department of
Agriculture (1982) and Reed (1988) that were used for the code acronyms.
bSymbols:*= Pteridophyes found in wetlands in the region according to Reed (1988); f= Pteridophytes from United States
Department of Agriculture (1982) not listed in Reed (1988); != Pteridophytes and non-vascular plants (macroalgae, mosses, or
liverworts) with artificial codes made up specifically for EMAP data.
life history codes: NP=native perennial; NA=native annual or biennial; IP=introduced or adventive perennial; IA=introduced or
adventive annual or biennial; -= Not applicable or unknown.
226
-------
ANOVA tables3.
Appendix 6-2
SV is source of variation, df is degrees of freedom, MS is
mean square, F is the F statistic, and P is the p-value.
Area of
Area of
Area of
Area of
Area of
low-prairie zone
SV
H
Error"
Y
Y*H
Error*
wet -meadow zone
SV
H
Error"
Y
Y*H
Error"
shallow-marsh zone
SV
H
Error"
Y
Y*H
Error1"
deep-marsh zone
SV
H
Error"
Y
Y*H
Error13
fen zone
SV
H
Error"
Y
Y*H
Error15
df
1
51
1
1
21
df
1
51
1
1
21
df
1
51
1
1
21
df
1
51
1
1
21
df
1
51
1
1
21
MS
0.00138
0.00115
<0. 00001
<0. 00001
0.00060
MS
31.1792
17.4515
0.0015
0.0268
0.0163
MS
26.3620
13.1675
3.1613
3.3065
3.5834
MS
29.6641
23.6985
0.0466
0.0855
0.5923
MS
0.0082
0.1258
0.0030
0.0030
0.0063
F
1.20
-
0.02
1.86
-
F
1.79
-
0.09
1.65
-
F
2.00
-
0.88
0.92
-
F
1.25
-
0.08
0.14
-
F
0.65
-
0.48
0.48
-
P
0.2775
_
0.8809
0.1866
P
0.1873
-
0.7652
0.2135
-
P
0.1632
-
0.3583
0.3477
-
P
0.2685
-
0.7820
0.7077
-
P
0.4221
-
0.4961
0.4961
-
Code to errors for zone area models: a=B(H); b=residual
227
-------
Appendix 6-2 (continued)
Area of communities
Full data model:
SV
H
Error*
Z
Z*H
Error*
Y
Y*H
Y*Z
Y*Z*H
Error0
Code to errors for full community
Reduced data model
SV
H
Y
Y*H
Error"
Z
Z*H
Z*Y
Z*Y*H
Error*
df
1
51
2
2
35
1
1
2
1
42
area
df
i
i
i
49
2
2
2
2
36
MS
54.63
44.31
1.64
2.78
5.04
0.52
0.04
1.04
n.e.b
0.45
model: a=B(H);
MS
16.27
9.68
8.54
41.36
0.20
0.56
0.35
0.36
4.81
F
1.23
-
0.33
0.55
-
1.16
0.09
2.30
n.t.°
-
b=Z*B(H); c=residual
F
0.39
0.23
0.21
-
0.04
0.12
0.07
o.oe
-
p
0.2721
0.7243
0.5796
0.2872
0.7718
0.1128
P
0.5335
0.6307
0.6516
0.9598
0.8898
0.9294
0.9277
Code to errors for reduced data model on community areas: a=B(H); b=residual
Water depth
Full data model:
SV
H
Error"
Z
Z*H
Error*
Y
Y*H
Y*Z
Y*Z*H
Error0
Code to errors for full water depth
Reduced data model
SV
H
Y
Y*H
Error*
Z
Z*H
Z*Y
Z*Y*H
Error*
df
1
51
2
2
35
1
1
2
1
42
model:
df
i
l
1
49
2
2
2
2
36
MS
60.57
300.69
4263.57
105.98
180.67
3492.81
96.66
376.01
n.e.*
136.93
a=B(H); b=Z*B(H);
MS
13.09
1568.28
58.28
241.62
2676.84
24.41
578.97
122.70
130.30
F
0.20
-
23.60
0.59
-
25.51
0.71
2.75
n.t.°
-
c=residual
F
0.05
6.49
0.24
_
20.54
0.19
4.44
0.94
_
P
0.6554
0.0001**
0.5616
0.0001**
0.4056
0.0757*
P
0.8169
0.0140**
0.6255
0.0001**
0.8299
0.0189**
0.3994
Code to errors for reduced data model on water depth: a=B(H); b=residual
228
-------
Appendix 6-2 (continued)
Percent dead vegetation
Full data model:
SV
H
Error"
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error0
Code to errors for full percent dead
Reduced data model
SV
H
Y
Y*H
Error"
Z
Z*H
Z*Y
Z*Y*H
Errorb
df
1
51
2
2
35
1
1
2
1
42
MS
123.76
93.33
213.80
1.43
22.64
59.38
28.01
15.83
n.e.b
9.02
vegetation model: a=B(H);
df
1
l
l
49
2
2
2
2
36
MS
216.72
78.57
33.17
94.09
128.83
23.82
7.71
3.12
19.00
P
1.33
™
9.44
0.06
6.58
3.11
1.75
n.t.°
-
b=Z*B(H);
F
2.30
0.84
0.35
_
6.78
1.25
0.41
0.16
_
P
0.2549
0.0005**
0.9389
0.0140**
0.0853*
0.1854
-
c=residual
p
0.1355
0.3653
0.5554
-
0.0032**
0.2976
0.6696
0.8490
_
Code to errors for reduced data model on percent dead vegetation: a=B(H); b=residual
Litter depth
Full data model:
SV
H
Error"
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error0
Code to errors for full litter depth
Reduced data model
SV
H
Y
Y*H
Error*
Z
Z*H
Z*Y
Z*Y*H
Error11
df
1
51
2
2
35
1
1
2
1
42
model:
df
l
1
1
49
2
2
2
2
36
MS
39.92
6.74
10.46
5.53
2.10
0.33
2.10
0.25
n.e.»
1.33
a=B(H); b=Z*B(H);
MS
9.25
11.87
35.43
6.39
7.62
3.92
2.87
9.18
1.95
F
5.92
-
4.97
2.58
-
0.25
1.58
0.19
n.t.°
-
c=residual
F
1.45
1.86
5.55
-
3.90
2.01
1.47
4.70
-
P
0.0185**
0.0126**
0.0902*
0.6217
0.2156
0.8296
P
0.2347
0.1791
0.0226**
0.0292**
0.1493
0.2442
0.0154**
Code to errors for reduced data model on litter depth: a=B(H); b=residual
229
-------
Appendix 6-2 (continued)
Percent unvegetated bottom
Full data model:
SV
H
Error"
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error"
Code to errors for full unvegetated
Reduced data model
SV
H
Y
Y*H
Error*
Z
Z*H
Z*Y
Z*Y*H
Error15
df
1
51
2
2
35
1
1
2
1
42
bottom
df
1
1
1
49
2
2
2
2
36
MS
6315.31
684.01
102.76
32.39
52.18
890.20
420.24
9.95
n.e.b
150.46
model: a=B(H);
MS
5581.24
2522.78
870.83
556.31
67.10
13.05
14.41
67.96
92.78
F
9.23
-
1.97
0.62
-
5.92
2.86
0.07
n.t."
-
b=Z*B(H); c=residual
F
10.03
4.53
1.57
-
0.72
0.14
0.16
0.73
-
P
0.0037**
-
0.1547
0.5434
-
0.0193**
0.0982*
0.9361
-
-
P
0.0026**
0.0383**
0.2168
-
0.4921
0.8692
0.8567
0.4878
-
Code to errors for reduced data model on percent unvegetated bottom: a=B(H); b=residual
Percent open water
Full data model:
SV
H
Error"
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error"
df
1
51
2
2
35
1
1
2
1
42
MS
182.62
744.61
2594.42
160.20
295.83
8313.44
1310.93
21.56
n.e.b
417.06
F
0.25
8.77
0.54
19.93
3.14
0.05
n.t."
P
0.6226
0.0008**
0.5867
0.0001**
0.0835*
0.9497
Code to errors for full percent open water model: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error*
Z
Z*H
Z*Y
Z*Y*H
Error*
df
1
1
1
49
2
2
2
2
36
MS
82.07
4187.18
116.04
817.85
1294.96
106.72
804.79
88.94
213.18
F
0.10
5.12
0.14
6.06
0.50
3.76
0.42
P
0.7528
0.0281**
0.7080
0.0054**
0.6111
0.0328**
0.6628
Code to errors for reduced data model on percent open water: a=B(H); b=residual
230
-------
.Appendix6-2 (continued)
Total plant species richness
Full data model:
SV
H
Error"
Z
Z*H
Error13
Y
Y*H
Y*Z
Y*Z*H
Error0
df
I
51
2
2
35
1
1
2
\
42
MS
578.90
98.25
1038.32
106.73
46.78
9.10
3.95
6.29
n.e.b
29.62
p
5.89
22.20
2.28
0.31
0.13
0.21
n.t.°
P
0.0188**
0.0001**
0.1171
0.5823
0.7169
0.8096
Code to errors for full model on total plant species richness: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV df MS
H 1 358.70
Y 1 186.27
Y*H 1 48.89
Error* 49 91.04
Z 2 832.00
Z*H 2 66.75
Z*Y 2 22.32
Z*Y*H 2 6.76
Error" 36 47.97
P
3.94
2.05
0.54
17.35
1.39
0.47
0.14
P
0.0528*
0.1589
0.4672
0.0001**
0.2617
0.6317
0.8690
Code to errors for reduced data model on total plant species richness: a=B(H); b=residual
Perennial plant species richness
Pull data model:
SV
H
Error"
Z
Z*H
Error*
Y
Y*H
Y*Z
Y*Z*H
Error"
df
1
51
2
2
35
1
1
2
1
42
MS
491.86
98.01
852.82
179.25
47.63
2.94
2.52
10.49
n.e.1
19.72
F
5.02
17.90
3.76
0.15
0.13
0.84
n.t.°
P
0.0295**
0.0001**
0.0331**
0.7014
0.7226
0.3657
Code to errors for full model on perennial plant species richness: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV df MS
H 1 303.80
Y 1 278.36
Y*H 1 36.71
Error" 49 87.99
Z 2 677.49
Z*H 2 121.43
Z*Y 2 25.92
Z*Y*H 2 7.56
Error" 36 41.46
F
3.45
3.16
0.42
16.34
2.93
0.63
0.18
P
0.0692*
0.0815*
0.5213
0.0001**
0.0663*
0.5408
0.8340
Code to errors for reduced data model on perennial plant species richness: a=B(H); b=residual
231
-------
Appendix A6-2 (continued)
Annual plant species richness
Full data model:
SV
H
Error"
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error0
df
1
51
2
2
35
1
1
2
1
42
MS
0.12
13.65
10.10
6.60
2.52
19.43
1.02
0.90
n.e."
3.70
F
0.07
4.00
2.62
5.26
0.28
0.24
n.t.°
P
0.9256
0.0272**
0.0872*
0.0270**
0.6027
0.7846
Code to errors for full model on annual plant species richness: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error"
Z
Z*H
Z*Y
Z*Y*H
Error"
df
1
1
1
49
2
2
2
2
36
MS
0.51
13.86
2.24
13.12
7.26
3.29
2.16
0.93
2.54
F
0.04
1.06
0.17
2.85
1.29
0.85
0.36
P
0.8438
0.3091
0.6809
0.0708*
0.2871
0.4355
0.6969
Code to errors for reduced data model on annual plant species richness: a=B(H); b=residual
Introduced plant species richness
Full data model:
SV
H
Error*
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error"
df
1
51
2
2
35
1
1
2
1
42
MS
6.03
7.10
108.45
1.97
3.65
0.01
0.29
0.69
n.e."
2.36
F
0.85
29.75
0.54
0.01
0.12
0.29
n.t.°
P
0.3613
0.0001**
0.5879
0.9793
0.7266
0.7480
Code to errors for full model on introduced plant species richness: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error*
Z
Z*H
Z*Y
Z*Y*H
Error"
d£
1
1
1
49
2
2
2
2
36
MS
4.71
3.76
1.57
6.00
66.33
0.70
4.69
0.01
4.06
F
0.79
0.63
0.26
16.35
0.17
1.15
0.01
P
0.3796
0.4325
0.6115
0.0001**
0.8420
0.3265
0.9975
Code to errors for reduced data model on introduced plant species richness: a=B(H); b=residual
232
-------
-Appendix 6-2 (continued)
Native plant species richness
Full data model:
SV
H
Error*
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error"
Code to errors for full model on
Reduced data model
SV
H
Y
Y*H
Error"
Z
Z*H
Z*Y
Z*Y*H
Error*1
df
1
51
2
2
35
1
1
2
1
42
native
df
1
l
l
49
2
2
2
2
36
MS
375.46
75.76
499.42
102.51
42.83
7.04
4.22
4.42
n.e.b
19.81
plant species richness:
MS
211.44
221.99
39.76
73.35
436.02
80.55
38.55
10.59
37.63
F
4.96
«
11.66
2.39
0.36
0.21
0.22
n.t.°
-
P
0.0305**
_
0.0001**
0.1061
0.5543
0.6468
0.8011
-
a=B(H); b=Z*B(H); c=residual
F
2.88
3.03
0.54
_
11.59
2.14
1.03
0.28
_
P
0.0959*
0.0882*
0.4651
_
0.0001**
0.1323
0.3657
0.7563
_
Code to errors for reduced data model on native plant species richness: a=B(H); b=residual
Species richness of introduced perennial plants
Full data model:
SV
H
Error*
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error0
df
1
51
2
2
35
1
1
2
1
42
MS
9.56
3.40
48.14
4.06
1.95
0.12
0.04
0.87
n.e.b
1.59
P
2.81
24.73
2.08
0.07
0.02
0.55
n.t.°
0.0999*
0.0001**
0.1395
0.7890
0.8764
0.5830
Code to errors for full model on species richness of introduced perennial plants: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error"
Z
Z*H
Z*Y
Z*Y*H
Error"
df
1
1
1
49
2
2
2
2
36
MS
7.87
0.24
1.75
2.79
30.12
1.27
1.51
0.33
2.42
F
2.82
0.09
0.63
12.45
0.52
0.63
0.14
P
0.0992*
0.7704
0.4319
0.0001**
0.5965
0.5396
0.8738
Code to errors for reduced data model on species richness of introduced perennial plants: a=B(H); b=residual
233
-------
Appendix 6-2 (continued)
Species richness of native perennial plants
Full data model:
SV
H
Error"
Z
Z*H
Error*
Y
Y*H
Y*Z
Y*Z*H
Error0
df
1
51
2
2
35
1
1
2
1
42
MS
364.30
85.14
505.16
129.37
44.42
1.89
1.93
5.38
n.e.b
14.75
F
4.28
11.37
2.91
0.13
0.13
0.36
n.t.°
P
0.0437**
0.0002**
0.0676*
0.7221
0.7192
0.6965
Code to errors for full model on species richness of native perennial plants: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error"
Z
Z*H
Z*Y
Z*Y*H
Error"
df
1
1
1
49
2
2
2
2
36
MS
213.87
262.25
22.43
78.66
427.95
98.96
39.72
5.31
35.41
F
2.72
3.33
0.29
12.08
2.79
1.12
0.15
P
0.1056
0.0740*
0.5958
0.0001**
0.0745*
0.3368
0.8613
Code to errors for reduced data model on species richness of native perennial plants: a=B(H); b=residual
Species richness of introduced annual plants
Full data model:
SV
H
Error*
Z
Z*H
Error1"
Y
Y*H
Y*Z
Y*Z*H
Error0
df
1
51
2
2
35
1
1
2
1
42
US
0.41
2.01
12.08
1.03
0.90
0.14
0.12
0.01
n.e.b
0.49
F
0.20
13.38
1.15
0.30
0.24
0.02
n.t.°
P
0.6553
0.0001**
0.3295
0.5889
0.6239
0.9780
Code to errors for full model on species richness of introduced annual plants: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error*
Z
Z*H
Z*Y
Z*Y*H
Error*
df
1
1
1
49
2
2
2
2
36
MS
0.40
5.90
0.01
1.92
7.09
0.09
1.14
0.38
0.77
F
0.21
3.07
0.01
9.22
0.12
1.48
0.50
P
0.6492
0.0859*
0.9593
0.0006**
0.8845
0.2410
0.6112
Code to errors for reduced data model on species richness of introduced annual plants: a=B(H); b=residual
234
-------
Appendix 6-2 (continued)
Species richness of native annual planta
Full data model:
SV
H
Error*
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error"
df
1
51
2
2
35
1
1
2
1
42
MS
0.08
7.40
1.25
2.42
1.49
16.23
0.44
0.75
n.e.b
2.58
p
0.01
0.84
1.63
6.28
0.17
0.29
P
0.9155
0.4395
0.2110
0.0162**
0.6815
0.7495
Code to errors for full model on species richness of native annual plants: a=B(H); b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error'
Z
Z*H
Z*Y
Z*Y*H
Error*
df
1
1
1
49
2
2
2
2
36
MS
0.01
1.68
2.46
7.44
1.81
2.53
0.20
2.40
1.99
F
0.01
0.23
0.33
0.91
1.27
0.10
1.21
P
0.9758
0.6372
0.5677
0.4111
0.2930
0.9037
0.3107
Code to errors for reduced data model on species richness of native annual plants: a=B(H); b=residual
Ratio of native annual planta to introduced annual plantaa
Full data model:
SV
H
Error*
Z
Z*H
Error"
Y
Y*H
Y*Z
Y*Z*H
Error"
df
1
51
2
2
35
1
1
2
1
42
MS
1.03
1.54
6.07
0.24
1.02
4.21
0.06
1.07
n.e.b
1.54
F
0.67
5.93
0.23
2.74
0.04
0.69
n.t."
P
0.4167
0.0061**
0.7935
0.1052
0.8463
0.5053
Code to errors for full model on ratio of native annual plants to introduced annual plants: a=B(H);
-b=Z*B(H); c=residual
Reduced data model
SV
H
Y
Y*H
Error"
Z
Z*H
Z*Y
Z*Y*H
Error"
df
1
1
1
49
2
2
2
2
36
MS
0.83
3.25
1.07
1.68
4.45
0.80
1.90
0.94
1.59
F
0.50
1.94
0.64
2.79
0.50
1.19
0.59
P
0.4841
0.1698
0.4283
0.0746*
0.6095
0.3163
0.5595
Code to errors for reduced data model on ratio of native annual plants to introduced annual plants: a=B(H);
b=residual
235
-------
Appendix 6-2 (continued)
Ratio of native perennial plants to introduced perennial plants"
Full data model:
SV df MS
H 1 14.26
Error" 51 15.67
Z 2 40.49
Z*H 2 6.60
Error" 35 9.80
Y 1 0.31
Y*H 1 2.19
Y*Z 2 0.18
Y*Z*H 1 n.e.b
Error0 42 5.00
F
0.91
4.13
0.67
0.06
0.44
0.04
n.t.c
P
0.3446
0.0245**
0.5164
0.8032
0.5120
0.9645
Code to errors for full model on ratio of native perennial plants to introduced perennial plants: a=B(H); b=Z*B(H);
c=residual
Reduced data model
SV df MS
H 1 1.15
Y 1 27.81
Y*H 1 0.22
Error" 49 15.01
Z 2 33.80
Z*H 2 23.97
Z*Y 2 26.28
Z*Y*H 2 14.50
Error" 36 8.24
F
0.08
1.85
0.01
4.10
2.91
3.19
1.76
P
0.7831
0.1796
0.9043
0.0248**
0.0674*
0.0531*
0.1865
Code to errors for reduced .data model on ratio of native perennial plants to introduced perennial plants: a=B(H);
b=residual
236
-------
.Appendix 6-2 (continued)
Percent of estimated basin watershed dominated by annual plants
SV df MS P P
H 1 102499.18 126.83 0.0001**
Error" 51 808.15
Y 1 23.48 0.52 0.4781
Y*H 1 23.48 0.52 0.4781
Error" 21 45.00
Percent of estimated basin watershed dominated by perennial plants
SV df MS P P
H 1 108962.61 162.44 0.001**
Error" 51 670.81
Y 1 23.48 0.08 0.7788
Y*H 1 23.48 0.08 0.7788
Error" 21 289.95
Percent of estimated basin watershed dominated by other than annual or perennial
SV
H
Error"
Y
Y*H
Error"
df
1
51
1
1
21
MS
261.99
185.11
1.16
1.16
2.41
P
1.42
-
0.48
0.48
-
P
0.2397
-
0.4958
0.4958
-
Code to errors for models of dominants of basin watersheds: a=B(H); b=residual
'Code to sources of variation: H=condition (health); B=basin; Z=zone; Y=year
bNot estimated
'Not tested
dRatio= (native annuals + 1)/(introduced annuals + 1) to accomodate zeros
6Ratio= (native perennials + 1)/(introduced perennials + 1) to accomodate zeros
237
-------
Appendix 6-3
Attributes of the 140 sampled communities, EMAP study, 1992-1993.
No. sampled quadrats per community
Attribute codes*
1
38
54
59
60
73
133
134
156
241
246
249
327
2
62
39
42
111
58
128
29
86
370
380
386
140
158
165
270
272
406
432
22
24
26
42
3
48
34
37
53
50
86
72
117
147
3_
2
1
1
2
2
1
3
1
1
3
2
3
2
3
1
2
2
1
3
2
1
2
3
2
1
1
3
1
2
1
3
2
4
WM
WM
DM
SM
WM
SM
WM
SM
WM
WM
DM
SM
WM
WM
WM
SM
WM
SM
WM
SM
WM
WM
SM
WM
WM
WM
DM
SM
WM
WM
SM
WM
WM
WM
WM
DM
SM
WM
WM
WM
WM
DM
SM
SM
WM
SM
WM
SM
WM
SM
WM
SM
1
1
4
3
1
2
2
1
2
1
1
3
2
1
1
1
2
2
1
2
1
1
2
1
1
1
2
1
1
1
3
2
1
1
2
1
1
1
1
1
2
3
4
1
2
1
1
1
2
2
1
2
1
2
1
3
1
2
2
C
C
I
X
I
I
G
G
I
I
I
I
I
I
G
C
C
I
I
I
I
I
I
I
C
I
C
C
C
T
C
I
I
I
C
G
G
G
G
G
I
I
I
I
C
C
C
C
I
I
G
G
M
M
I
C
I
C
I
I
7
CD
CT
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
CT
CD
NE
NE
NE
NE
NE
CD
NE
NE
CT
NE
NE
CT
CT
CD
CD
NE
NE
NE
CT
NE
NE
NE
NE
NE
NE
NE
NE
NE
CT
CT
CT
CT
NE
NE
NE
NE
NE
NE
NE
CT
NE
CT
NE
NE
1992
G"
5
5
5
5
5
5
5
5
5
t
f
f
f
.
t
.
.
t
t
^
t
.
t
.
,
.
.
(
t
,
5
5
5
5
5
,
.
.
,
f
t
.
f
f
f
S
s
5
5
t
.
.
.
f
t
5
5
5
5
5
5
.
.
.
.
.
f
.
f
t
,
r
s
5
.
4
,
5
.
,
.
S
5
.
5
5
5
f
.
,
.
,
5
5
5
5
5
5
5
5
5
5
t
t
f
t
,
t
.
,
^
m
1993
G
.
5
5
5
5
.
.
.
»
5
5
.
.
.
,
»
.
.
.
,
.
,
,
.
,
*
5
5
5
5
5
,
.
.
.
f
f
.
f
f
.
t
.
t
r
f
f
f
f
,
f
f
*
.
.
.
.
.
5
5
5
5
*
»
.
.
5
5
5
,
5
5
5
.
»
5
5
5
5
.
,
.
»
,
.
*
.
t
f
t
,
t
f
t
t
f
t
f
5
5
S
5
5
5
238
-------
Appendix 6-3 (continued)
1
363
374
396
407
442
498
Attrib
2
22
58
65
100
225
272
106
107
130
67
109
168
93
260
261
281
295
301
146
227
277
ute c
3^
3
1
1
3
2
2
3
1
3
2
1
3
2
1
1
2
3
3
3
1
2
3 odes"
4
WM
DM
VIM
SM
VIM
WM
DM
SM
WM
DM
SM
VIM
WM
WM
WM
SM
WM
WM
WM
DM
SM
WM
SM
WM
WM
WM
SM
WM
DM
SM
WM
DM
SM
WM
DM
WM
WM
DM
SM
WM
JE.
i
3
1
2
1
1
3
2
1
3
4
2
3
1
1
2
1
1
1
2
1
1
3
2
1
2
1
1
1
2
1
3
2
1
3
2
1
3
1
1
3
2
1
£
I
I
I
G
6
G
G
G
G
G
G
G
G
G
G
G
G
M
G
G
G
I
G
G
G
I
I
C
I
C
I
I
I
T
I
I
I
I
I
I
I
I
I
I
I
7_
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
NE
ND
NE
ND
ND
NE
NE
NE
NE
NE
NE
NE
CT
CD
CT
CD
NE
NE
NE
ND
NE
ND
NE
NE
ND
NE
ND
NE
NE
NE
NE
NE
NE
NE
1992
G" I
.
5
5
5
5
5
5
5
5
5
t
5
t
5
5
5
5
5
5
5
t
t
t
f
t
S
5
t
I
. ,
C
1
1
,
5
.
5
.
5
5
,
5
.
t
.
,
,
,
•
1
G
5
5
5
5
5
5
5
5
5
5
5
5
5
5
\
<
)
5
5
5
5
5
5
5
5
»
5
t
5
5
t
5
t
5
5
5
5
5
5
5
993
P
5
f
•
•
t
>
,
%
.
5
,
5
,
5
5
.
,
,
f
t
f
t
.
.
.
f
f
,
.
.
*
"Attribute codes: Col 1 = Plot No.; Col 2 = Basin No.; Col 3 = Basin Class (1=temporarily-flooded; 2=seasonally-flooded;
3=semipermanently-flooded); Col 4 = Zone (WM=wet- meadow; SM=shallow-marsh; DM=deep-marsh); Col 5 = Community
number within basin; Col 6 = Land-use (l=ldle; M=Mowed; C=Cultivated; G=GRAZED); Col 7 = Phase NE=Normal Emergent;
CT=Cropland Tillage; CD=Cropland Drawdown; ND=Natural Drawdown.
bG=Basin in good-condition watershed.
°P=Basin in poor-condition watershed.
239
-------
APPENDIX 7-1
Appendix 7-1. EMAP Wetland Soil Classifications by John Freeland, 1992-93.
EM-38
Plot
38
38
38
38
38
54
54
54
54
54
59
59
59
59
59
59
59
59
59
59
59
59
59
59
59
WL
62
62
62
62
62
39
39
39
39
39
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
Health
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Zone
WM
WM
^^tfwwwwwwwwwwwwww
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
sa
SM
Comm.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
3
3
3
3
3
Year
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
Quad
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Soil Classification
Aerie Calciaquoll
Aerie Calciaquoll
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^•^^ • i •• n»i^»^^^^^»»**»»-p
Aerie Calciaquoll
Cumuli c Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumuli c (Calc) Endoaquoll
Typi c (Calc) Endoaquol 1
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic Calciaquoll
Typic (Calc) Endoaquoll
V
38
42
52
50
42
190
170
180
180
150
60
72
88
68
82
50
70
70
54
68
80
90
64
86
100
H
28
37
^^^^^^^^•^•^•^^^^^
44
40
42
215
220
270
230
240
62
70
82
62
50
46
66
66
54
54
78
88
60
80
100
-------
59
59
59
59
59
59
59
59
59
•••^•^••m^V^^^^^BWM
59
59
59
59
59
59
60
60
60
60
60
60
60
60
60
60
60
60
60
60
42
42
42
42
42
111
111
111
111
111
111
111
111
111
111
58
58
58
58
58
58
58
58
58
58
128
128
128
128
Low
Low
Low
Low
LOW
High
High
High
High
^^^^^W«*MV^Hri^^^— ^^V^^H^W
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
^^^^•^^^P^IVWHWHIBaHI^H^^IVB
High
High
High
High
DM
DM
DM
DM
DM
WM
WM
VIM
VIM
VIM
SM
SM
SM
SM
SM
WM
WM
VIM
WM
WM
SM
SM
SM
SM
SM
mtrm-mtm^^^m^^^^^^^ nmt' ' I
WM
WM
WM
WM
4
4
4
4
4
1
1
1
1
1
2
2
2
2
2
1
1
1
1
1
2
2
2
2
2
1
1
1
1
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
1
2
3
4
5
1
2
3
4
^^^V*^^— ^^^^V^BWM
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
Cumuli c (Gale) Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typio Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Aerie Fluvaquent
. — • i i.i ii .1.1 -i. ... — -.-- -— . -^ ^ — i .... — .
Aerie Fluvaquent
Typic Calciaquoll
Aerie Calciaquoll
Aerie Calciaquoll
Aerie Calciaquoll
Aerie Calciaquoll
Typic Calciaquoll
Cumuli c (Calc) Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
M«^— «---M««-PB-BBBBBBBBV^ta^^^— ^••••^^
Aerie Calciaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
115
76
92
90
96
120
70
74
70
rtVmH^^^MHM^^H^^^H
52
90
70
68
80
72
76
48
46
50
52
54
66
34
86
84
^^^M^B^B^^^^^h^^pBIV^H
20
30
62
54
100
70
84
80
110
90
60
68
60
50
62
80
70
62
44
30
40
40
46
45
49
58
30
70
56
18
24
38
38
-------
<•
60
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
73
128
29
29
29
29
29
29
29
••— ^^^— ^— — ««.^^^^^
29
29
29
29
29
29
29
29
29
29
29
29
29
29
29
29
29
29
29
29
29
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
WM
MM
WM
VIM
WM
WM
WM
WM
p**^-^— I— ^MBVta^AB-fe^B
WM
SM
SM
SM
SM
SM
SM
SM
SM
DM
DM
DM
DM
DM
DM
DM
DM
DM
DM
DM
DM
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
4
4
4
4
92
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
5
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
Cumuli c (Calc) Endoacjuoll
Typic Psammaguent
Typic Psammaguent
Typic Psammaguent
Typic Psammaguent
Typic Psammaquent
Typic Psammaquent
Typic Psammaguent
Typic Psammaguent
Typic Psammaguent
Typic Psammaguent
Typic Psammaguent
Typic Psammaquent
Typic Psammaguent
Typic Psammaguent
Typic Psammaguent
Typic Psammaguent
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
52
20
23
48
40
44
^^^^^mm—^^mummmm^mmmmm
46
80
140
125
105
flood
flood
flood
flood
flood
flood
flood
flood
flood
34
18
22
32
30
36
W^-MMM-M^^>B>M»»»^^^^^»>"
30
115
125
120
95
-------
1
73
73
73
73
73
73
73
73
73
73
73
73
133
133
133
133
133
133
133
133
133
133
133
133
133
133
•^^^^^HHHHHHHHHPB-^^
133
133
133
29
29
29
29
86
86
86
86
86
86
86
86
370
370
370
370
370
370
380
380
380
380
380
380
386
386
386
386
386
High
High
High
High
High
High
High
High
High
High
High
High
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
High
^^^B-WVBV^^—^O^^^^^^^^^—
High
High
High
DM
DM
DM
DM
WM
WM
VIM
WM
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
WM
WM
WM
SM
SM
SM
WM
WM
WM
SM
SM
4
4
4
4
1
1
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
2
2
2
1
1
1
2
2
92
93
93
93
92
92
92
92
92
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
5
1
2
4
1
2
3
4
5
1
2
4
1
2
4
1
2
4
1
2
4
1
2
4
1
2
4
1
2
Typio Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Aerie Calciaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
flood
28
40
46
46
58
36
54
56
52
62
-------
to
133
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
13d
134
134
134
134
134
134
134
134
134
134
386
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
158
158
158
165
165
165
165
165
165
165
165
165
High
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
LOW
LOW
Low
Low
Low
Low
Low
Low
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
WM
WM
WM
SM
SM
SM
SM
SM
WM
WM
WM
SM
2
1
1
1
1
1
2
2
2
2
2
1
1
1
2
2
2
1
1
1
1
1
1
1
1
1
1
1
2
93
92
92
92
92
92
92
92
92
92
92
93
93
93
93
93
93
93
93
93
92
92
92
92
92
93
93
93
93
4
1
2
3
4
5
1
2
3
4
5
1
2
4
1
2
4
1
2
3
1
2
3
4
5
1
2
4
1
Typic Argiaquoll
Typic Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Endoaquoll
Typic Calciaquoll
Typic Endoaquoll
Typic Endoaquoll
Cumulic (calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Typic Argiaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumulic (Calc) Endoaquoll
Cumulic (calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Typic Calciaquoll
Aerie Calciaquoll
Typic Calciaquoll
Cumulic (Calc) Endoaquoll
42
30
40
52
40
75
78
70
70
74
40
42
40
44
42
28
20
26
34
22
60
60
56
56
60
26
28
24
30
28
-------
ro
&
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
134
165
165
270
270
270
270
270
270
270
270
272
272
272
272
272
406
406
406
406
406
406
406
406
406
406
406
406
406
406
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
SM
SM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
SM
SM
SM
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
92
92
92
92
92
93
93
93
92
92
92
92
92
93
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
3
4
5
1
2
4
1
2
3
4
5
1
Cumuli c (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumulic Calciaquoll
Cumulic Calciaquoll
Cumulic Calciaquoll
Cumulic Calciaquoll
Cumulic Calciaquoll
Typic Calciaquoll
52
48
52
50
50
68
86
84
76
78
48
45
42
40
35
flood
flood
flood
flood
flood
48
50
50
50
50
54
78
66
60
60
32
30
28
26
20
-------
134
134
134
134
134
134
134
134
134
134
134
134
134
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
406
406
406
406
406
432
432
432
432
432
432
432
432
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
SM
SM
DM
DM
DM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
SM
SM
SM
SM
SM
2
2
3
3
3
1
1
1
1
1
1
i
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
93
93
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
2
4
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
3
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Aerie Calciaquoll
Typic Calciaquoll
Aerie Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Argiaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Typic Calciaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
46
3
10
32
30
34
36
46
59
49
76
86
92
90
88
30
26
22
26
28
24
34
34
44
35
53
58
62
70
62
-------
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
156
241
241
241
241
241
24
24
24
24
24
24
24
24
26
26
26
26
26
26
26
26
42
42
42
42
42
42
42
42
3
3
3
3
3
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
Low
Low
Low
Low
Low
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
DM
DM
DM
DM
DM
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
1
2
3
4
B
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Epiaquent
Typic Epiaquent
Typic Epiaquent
Typic Epiaquent
Typic Epiaquent
Typic Epiaquent
Typic Epiaquent
Typic Epiaquent
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cunvulic (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
22
40
38
49
56
54
49
76
74
86
52
66
86
72
54
flood
flood
flood
flood
flood
18
38
24
35
40
44
40
52
60
62
30
36
44
50
40
-------
ro
£t
00
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
241
246
246
246
246
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
48
48
48
48
48
48
48
48
48
48
34
34
34
34
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
SM
SM
SM
SM
SM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
2
2
2
2
2
3
3
3
3
3
4
4
4
4
4
1
1
1
1
1
2
2
2
2
2
1
1
1
1
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
Cumuli o (calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Aerie Calciaquoll
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Aerie Calciaquoll
Aerie Calciaquoll
Aerie Calciaquoll
Aerie Calciaquoll
Typic Calciaquoll
Cumulic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Endoaquoll
Typic Endoaquoll
flood
flood
flood
flood
flood
flood
flood
flood
flood
flood
flood
flood
flood
flood
flood
95
300
98
98
94
94
flood
flood
flood
flood
60
56
56
49
90
340
96
96
92
90
50
45
43
40
-------
ro
246
246
246
246
246
246
246
246
246
246
246
246
246
246
246
246
249
249
249
249
249
249
249
249
249
249
249
249
249
34
37
37
37
37
37
S3
53
53
53
53
53
53
53
53
53
50
50
50
50
50
50
50
50
50
50
86
86
86
Low
LOW
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
WM
WH
WM
WM
WM
WM
DM
DM
DM
DM
DM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
WM
WM
WM
WM
WM
WM
WM
WM
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
92
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
Typic Calciaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Typic Endoaquoll
Cumuli c Endoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
62
55
58
57
58
54
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
46
50
50
50
SO
40
-------
ro
s
249
249
249
249
249
249
249
327
327
327
327
327
327
327
327
327
327
327
327
327
327
327
327
327
327
327
327
327
363
86
86
86
86
86
86
86
72
72
72
72
72
72
117
117
117
117
117
117
117
117
117
147
147
^^^H^WBA*^^HP^ta^^H^B1
147
147
147
147
22
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
High
WM
WM
SM
SM
SM
SM
SM
WM
WM
WM
SM
SM
SM
WM
WM
WM
WM
WM
WM
SM
SM
SM
WM
WM
WM
SM
SM
SM
WM
1
1
2
2
2
2
2
1
1
1
2
2
2
1
1
1
2
2
2
3
3
3
1
— r
2
2
2
1
92
92
92
92
92
92
92
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
92
4
5
1
2
3
4
5
1
2
4
i
2
4
1
2
4
1
2
4
i
2
4
1
2
4
1
2
4
1
Cumuli c Endoaguoll
Cumuli c Endoaguoll
Cumuli c Endoaguoll
Cumulic Endoaguoll
Cumuli c Endoaguoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Typic Calciaguoll
Typic Calciaguoll
Typic Calciaguoll
Cumulic Endoaguoll
Cumulic Endoaguoll
Cumulic Endoaguoll
Typic Calciaguoll
Typic Epiaguoll
Typic Epiaguoll
Typic Argiaguoll
Typic Argiaguoll
Typic Epiaguoll
Typic Calciaguoll
Typic Calciaguoll
Typic Calciaguoll
Typic Endoaguoll
Typic Endoaguoll
Typic Endoaguoll
Typic Argiaguoll
Typic Argiaguoll
Typic Argiaguoll
Typic Endoaquoll
Flood
Flood
Flood
Flood
Flood
Flood
Flood
*W«V^V1^^«WM^^^^^^^
Flood
-------
Ol
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
363
^^ p^^^^^^^fc^—
363
363
363
363
22
22
22
22
22
22
22
22
22
22
22
22
22
22
22
58
58
58
58
58
58
58
58
58
58
^V^^^^^^^^^^^B^^^^—
58
58
58
58
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
WM
WM
WM
WM
WM
WM
WM
DM
DM
DM
DM
DM
DM
DM
DM
WM
WM
WM
WM
WM
SM
SM
SM
SM
SM
WM
WM
WM
WM
1
1
1
1
1
1
1
3
3
3
3
3
3
3
3
1
1
1
1
1
2
2
2
2
2
1
1
1
2
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
92
92
92
92
92
93
93
93
93
2
3
4
5
1
2
4
1
2
3
4
5
1
2
3
1
2
3
4
5
1
2
3
4
5
1
2
4
1
Typic Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Cumuli c Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Typic Calciaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Typic Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Aaric Calciaquoll
Aerie Calciaquoll
Typic Calciaquoll
Typic Argiaquoll
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
140
135
130
100
120
Flood
Flood
Flood
Flood
Flood
160
140
140
120
110
^^^w-p«^^^^_«.^^^^_
-------
ro
0!
ro
363
363
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
58
58
65
65
65
65
65
65
65
65
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
WM
WM
VIM
WM
TIM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
SM
SM
SM
SM
SM
DM
DM
DM
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
3
3
3
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
Typic Argiaquoll
Typic Argiaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Calciaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cxumilic (Calc) Endoaquoll
Typic Calciaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Cumulic Endoaquoll
Cumuli c Endoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
-------
ro
en
Co
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
100
100
100
100
100
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
225
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
DM
DM
DM
DM
DM
VIM
WM
VIM
VIM
VIM
VIM
VIM
VIM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
SM
DM
DM
DM
DM
DM
3
3
3
3
3
1
1
1
1
1
1
1
1
2
2'
2
2
2
2
2
2
3
3
3
3
3
3
3
3
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
93
93
93
92
92
92
92
92
4
5
1
2
3
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
4
1
2
3
4
5
Typic Endoaquoll
Typic Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumulio (Calc) Endoaquoll
Typic Fluvaquent
Aerie Calciaquoll
Typic Calciaquoll
Cumulic Endoaquoll
Cumulic (Calc) Endoaquent
Typic Endoaquoll
Typic Endoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
-------
ro
2
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
374
396
396
396
396
396
396
396
396
396
396
225
225
225
272
272
272
272
272
272
272
272
272
272
272
272
272
272
272
272
106
106
106
106
106
107
107
107
107
107
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
DM
DM
DM
WM
WM
WM
WM
WM
VIM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
4
4
4
1
1
1
1
1
2
2
2
2
2
1
1
1
2
2
2
1
1
1
1
1
1
1
1
1
1
93
93
93
92
92
92
92
92
92
92
92
92
92
93
93
93
93
93
93
92
92
92
92
92
92
92
92
92
92
1
2
4
1
2
3
4
5
1
2
3
4
5
1
2
4
1
2
4
1
2
3
4
5
1
2
3
4
5
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Cumulic (Calc) Endoaquoll
Cumuli o (Calc) Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Pachic Udic Haploboroll
Typic Haploboroll
Pachic Udic Haploboroll
Typic Haploboroll
Typic Haploboroll
Typic Haploboroll
Typic Haploboroll
Typic Haploboroll
Typic Haploboroll
Typic Haploboroll
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
42
28
37
29
29
25
23
24
23
22
30
24
32
23
20
21
21
20
18
19
-------
Ul
396
396
396
396
396
396
396
396
396
396
407
407
407
407
407
407
407
407
407
407
407
407
407
407
407
442
442
442
442
130
130
130
130
130
130
130
130
130
130
67
67
67
109
109
109
168
168
168
168
168
168
168
168
168
93
93
93
93
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
SM
SM
SM
SM
SM
mi
WM
TIM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
DM
DM
DM
WM
WM
WM
WM
1
1
1
1
1
2
2
2
2
2
1
1
1
1
1
1
1
1
1
2
2
2
3
3
3
1
1
1
1
92
92
92
92
92
92
92
92
92
92
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
92
92
92
92
1
2
3
4
5
1
2
3
4
5
1
2
4
1
2
4
1
2
4
1
2
4
1
2
4
1
2
3
4
Cumuli c Endoaquoll
Typic Bndoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Fsammaquent
Typic Fsammaquent
Typic Epiaquent
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Mollic Fluvaquent
Typic Fluvaquent
Mollic Fluvaquent
Typic Fluvaquent
48
50
51
58
46
27
35
35
34
48
30
24
38
46
34
39
38
42
36
16
26
25
21
34
16
16
20
28
-------
ro
01
O>
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
93
93
93
93
93
93
93
93
93
93
93
93
260
260
260
260
260
260
260
260
261
261
261
261
261
261
261
261
281
High
High
High
High
High
High
High
High
High
High
High
High
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
LOW
LOW
Low
Low
WM
WM
WM
WM
SM
SM
SM
SM
SM
SM
SM
SM
VIM
WM
VIM
VIM
VIM
VIM
VIM
VIM
WM
WM
WM
WM
WM
WM
WM
WM
WM
1
1
1
1
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
Typio Fluvaquent
Typic Argiaquoll
Typio Fluvaquent
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Epiaquoll
Typic Epiaquoll
Typic Epiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Epiaquent
Typic Epiaquent
Typic Epiaquent
Typic Argiaquoll
26
40
53
46
50
46
34
46
64
70
64
60
62
58
50
44
42
16
22
36
30
30
28
28
32
50
50
44
42
50
44
34
32
20
-------
en
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
281
281
281
281
281
281
281
281
281
281
281
281
281
281
281
295
295
295
295
295
295
295
295
295
295
295
295
295
295
Low
LOW
LOW
Low
•0^— ^rtM^HM-M^MV^B^MM^— ^^B«
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
SM
SM
SM
SM
SM
WM
WM
WM
WM
WM
WM
WM
WM
SM
SM
SM
SM
SM
SM
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
2
2
2
2
2
2
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
Typio Argiaquoll
Typic Argiaquoll
Typio Argiaquoll
Typic Argiaquoll
Typic Epiaquoll
Typic Epiaquoll
Typic Epiaquoll
Cumuli c Endoaquoll
Cumuli c Endoaquoll
Cumulic Endoaquoll
Cumuli c Endoaquoll
Cumulic Endoaquoll
Typic Epiaquoll
Typic Epiaquoll
Typic Epiaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Epiaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Epiaquoll
54
48
40
38
^^^^MBM^VBMW^H^^—
64
68
68
68
56
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
Flood
30
26
20
20
40
40
43
38
30
-------
N>
Ol
00
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
442
295
295
295
295
295
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
301
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
LOW
Low
Low
Low
Low
Low
Low
Low
SM
SM
DM
DM
DM
WM
WM
WM
WM
KM
WM
WM
WM
SM
SM
SM
SM
SM
SM
SM
SM
DM
DM
DM
DM
DM
DM
DM
DM
2
2
3
3
3
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
93
93
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
92
92
92
92
92
93
93
93
2
4
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
1
2
3
4
5
1
2
4
Typio Argiaquoll
Typic Calciaquoll
Typic Endoaquoll
Typio Endoaquoll
Typic Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Cumuli c (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumuli c (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
Cumulic (Calc) Endoaquoll
420
420
340
340
440
440
340
430
400
340
Flood
Flood
Flood
Flood
Flood
300
300
220
240
360
320
220
340
320
270
-------
CO
498
498
498
498
498
498
498
498
498
498
498
498
498
498
498
498
498
498
498
498
498
146
146
146
146
146
146
146
146
146
227
227
227
277
277
277
277
277
277
277
277
277
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
High
WM
WM
WM
SH
SH
SM
DM
DM
DM
WM
WM
mi
WM
WM
WM
SM
SM
SM
DM
DM
DM
1
1
1
2
2
2
3
3
3
1
1
1
1
1
1
2
2
2
3
3
3
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
93
1
2
4
1
2
4
1
2
4
1
2
4
1
2
4
1
2
4
1
2
4
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaguoll
Typic Endoaquoll
Typic Endoaquoll
Typic Endoaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
Typic Argiaquoll
-------
Appendix 7-2
Soil Characterization Data from CWLSA 1992. John Fr eel and and Jim Richardson, investigators. Abbreviations: WL=wetland,
Tran=transect, OM=organic matter, DSD=dry soil density.
WL
PI
PI
PI
PI
PI
PI
PI
PI
Pi
PI
PI
PI
PI
PI
PI
PI
PI
PI
Pi
PI
Pi
PI
Tran
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
Haalt
h
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Zone
wm
wm
wm
wm
am
am
sm
am
dm
dm
din
dm
wm
wm
wm
wm
am
sm
ant
am
dm
dm
Midpt .
Depth
(in)
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
N03 g/m3
4.3
2.7
1.8
i.e
4.7
1.6
1.8
3.7
29.2
7.8
4.5
2.3
4.2
1.6
1.4
1.2
5.2
2.7
1.1
0.8
10
4.7
NaHCOS
ext. P
g/m3
7.6
5.3
2.9
2.4
4.7
2.4
2.4
2.9
7.6
5.9
4.7
2.9
7.1
4.7
2.9
2.4
5.9
4.1
2.9
2.4
5.9
4.1
% OH
4.0
3.7
2.2
1.5
2.5
2.1
1.7
1.0
5.2
6.4
6.0
3.4
4.4
3.0
3.3
2.9
5.4
2.7
2.2
1.6
3.6
3.2
pH
7.5
7.8
8
8.1
7.6
8.2
8.1
8.1
7.5
7.5
7.6
7.7
7.4
7.6
7.9
7.9
7.4
7.5
7.8
7.8
7.3
7.3
EC
micro-mhos
540
1650
3500
3500
2500
3400
3500
3300
2500
2700
2800
2400
540
260
560
1700
1800
1800
1480
1430
1420
1790
DSD
g/cm3
0.9
1.21
1.15
1.2
1.06
1.17
1.08
1.03
0.5
0.84
0.94
1.11
1.08
1.36
1.17
1.09
0.88
1.17
1.33
1.42
0.68
0.91
%
Sand
63.0
41.6
15.9
21.6
60.4
42.2
15.8
5.8
43.1
27.5
25.9
36.5
87.3
91.5
48.3
41.7
72.5
78.0
93.5
94.9
62.5
81.0
%
Silt
27.1
35.5
36.8
33.7
39.6
24.0
33.9
43.4
52.3
64.6
55.6
39.0
10.1
6.3
23.5
29.7
20.9
15.5
5.9
3.8
35.5
16.3
% Clay
9.8
23.0
47.3
44.7
0.0
33.8
50.3
50.8
4.6
7.9
18.5
24.6
2.6
2.2
28.2
28.5
6.5
6.5
0.6
1.3
2.0
2.7
-------
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
Pi
PI
PI
PI
PI
PI
PI
PI
2
2
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
H
B
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
dm
Cult
win
win
vnn
wm
sm
sm
sm
sm
dm
dm
dm
dm
wm
wm
wm
wm
sm
sm
sm
sm
dm
CUR
<3ni
dm
wm
wm
wm
32.5
52.5
7.5
22.5
••(-•^^•••••^••^^•qta^A*!^^—
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
4
1.5
2.4
1.2
^•VB^OHW-^IMIBIIV^Hf^^
0.9
1.2
6.8
2.4
1.1
0.7
7
2.9
1.3
1.4
2.6
1
0.7
0.8
2.4
2.7
2.1
0.9
5
2.2
1
1
1.9
1.1
0.9
3.5
3.5
5.9
3.5
2.9
2.4
4.1
1.8
1.8
2.4
5.9
4.7
2.4
3.5
5.9
4.1
3.5
3.5
5.3
4.1
2.9
2.4
3.5
2.9
2.4
2.9
4.7
4.1
2.9
3.9
3.2
2.9
2.2
1.8
1.8
1.8
2.2
2.2
1.8
3.7
3.0
4.0
3.3
2.2
2.6
2.9
5.9
2.6
2.9
2.6
2.9
2.6
2.6
2.6
2.6
2.6
2.2
2.6
7.4
7.6
7.6
7.9
^M^**««"«*l*-^h"W««w
8
8
7.4
7.5
7.8
8
7.4
7.6
7.6
7.5
7.5
7.6
7.9
7.9
7.3
7.6
7.9
7.9
7.3
7.4
7.6
7.7
7.7
7.9
8.1
1970
1700
300
230
^— ••— «^— l^—fc— «l«™l— ^Kl«^— ™^
350
1000
1820
1950
1900
1650
520
500
520
1400
250
140
130
390
200
160
500
1960
450
350
510
460
250
260
1660
1.07
1.25
1.23
1.34
1.21
1.19
0.81
1.03
1.1
1.23
0.68
1.01
1.12
1.1
1.14
1.42
1.37
1.31
1.21
1.44
1.44
1.3
0.89
1.1
1.31
1.28
1.2
1.23
1.15
66.8
77.2
87.5
89.7
80.6
73.9
84.4
81.2
89.7
92.8
70.1
72.7
80.6
64.7
89.6
85.1
94.9
74.6
88.6
93.7
97.5
81.9
79.3
85.2
92.3
86.4
91.2
88.4
51.4
29.9
14.9
14.1
5.7
•^•^—•^•••••••^••p^— •——••••—
12.2
14.4
15.6
14.1
6.4
5.1
26.2
14.2
11.6
22.2
7.8
11.5
4.4
14.8
11.5
5.0
2.5
12.8
20.1
13.4
6.4
8.9
6.9
8.4
25.7
3.3
8.0
0.0
4.6
7.2
11.7
0.0
4.7
3.9
2.1
4.5
13.1
7.8
13.1
2.6
3.4
0.6
10.6
0.6
1.3
0.0
5.3
0.7
1.4
1.3
4.7
1.9
3.2
22.9
-------
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
5
5
5
5
5
B
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
7
7
7
7
7
7
7
7
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
vim
sm
am
sm
sm
dm
dm
dm
dm
win
wm
wm
wm
sm
sm
sm
sm
dm
dm
dm
dm
wm
wm
wm
wm
sm
sm
sm
sm
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
1.1
0.9
1.1
!
1
11
2.7
1.2
1.1
2.2
1
0.9
0.8
3.4
1.1
1.1
1
8.8
5.4
3.2
1.8
5.1
1.1
1
1.2
6.6
1.8
0.9
1.2
1.8
4.7
4.1
2.9
1.8
4.1
2.4
1.8
1.8
4.7
3.5
2.9
2.4
4.1
2.4
1.8
1.8
2.9
2.9
2.9
2.4
12.9
3.5
2.4
2.4
7.6
2.4
1.8
1.8
2.6
2.9
2.6
4.8
2.2
1.8
2.2
4.0
3.3
3.3
2.6
2.9
1.8
2.9
2.9
3.3
1.8
3.3
3.3
3.7
3.3
4.4
2.2
2.6
2.2
6.7
2.2
2.6
2.2
8.1
7.5
7.9
8
8
7.2
7.6
7.9
7.9
7.7
8.1
8.1
8.2
7.6
8
8.1
8
7.3
7.5
7.6
7.7
7.5
7.6
7.5
7.8
7.5
7.9
8.2
8
3000
250
250
1100
2000
2200
2100
1900
1900
240
270
1130
2400
460
1200
2100
2300
1900
2000
1900
1700
390
150
170
700
1180
1800
1900
1850
1.13
1.13
1.29
1.25
1.13
0.72
1.06
1.27
1.28
1.05
1.29
1.18
1.21
1.04
1.26
1.18
1.16
0.75
0.89
1.03
1.24
1
1.29
1.33
1.24
0.87
1.28
1.31
1.31
45.9
87.4
87.7
71.3
47.8
55.9
72.7
72.7
70.0
84.5
82.4
49.6
47.8
64.7
68.7
48.6
45.9
43.8
52.7
59.0
74.6
80.8
95.0
93.6
56.5
74.1
85.1
83.3
62.6
28.6
12.6
9.6
18.2
25.9
40.8
18.8
17.5
17.5
12.8
11.0
30.0
31.2
22.9
18.9
30.4
32.3
51.7
38.0
29.9
15.5
IS. 3
3.7
5.8
23.2
25.9
10.9
10.8
17.1
25.5
0.0
2.7
10.5
26.3
3.2
8.5
9.8
12.4
2.6
6.6
20.3
21.1
12.4
12.4
21.0
21.8
4.5
9.3
11.1
9.9
3.9
1.3
0.6
20.3
0.0
4.0
5.9
20.3
-------
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
7
7
7
7
8
8
8
8
8
8
8
8
8
8
8
8
9
9
9
9
9
9
9
9
10
10
10
10
10
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
B
H
H
H
H
H
H
H
H
H
H
dm
dm
Qin
dm
WIU
win
wm
wm
sm
sm
sm
sm
din
dm
ctm
dm
wm
vim
wm
wm
dm
Ulll
Cull
elm
wm
wm
wm
wm
sm
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
11.3
4.5
1.6
1.2
3.2
1.9
1.1
1.1
6.3
2.1
1.5
1.5
S.2
2.5
1.9
1.7
2.6
1.4
1.2
1.1
5.9
1.9
1.2
1.3
3.1
1.9
1.5
1.4
4.6
8.8
5.3
2.9
3.5
5.9
3.5
1.8
1.8
7.6
3.5
2.4
1.8
6.5
3.5
2.4
2.4
3.5
2.4
1.8
1.8
5.9
2.9
1.8
1.8
5.3
2.9
1.8
1.8
6.5
4.4
2.9
2.9
3.0
3.0
2.6
2.6
2.6
4.4
3.6
2.9
2.6
3.7
4.8
3.7
2.9
3.7
2.9
2.2
2.2
2.S
3.3
3.3
2.2
4.0
2.6
2.2
2.2
3.7
7.2
7.5
7.7
7.7
7.8
7.9
8.1
8.2
7.5
7.8
8
8
7.3
7.7
7.6
7.7
7.8
7.1
8.2
8.4
7.4
7.7
7.8
7.7
7.8
7.9
7.9
7.9
7.6
900
600
1070
1450
300
310
1120
2000
430
490
1000
1450
1960
2000
2000
2000
340
1080
2500
2700
700
710
1650
2000
740
1520
2200
2500
570
0.72
1.03
1.19
1.24
1
1.11
1.24
1.29
0.84
1.16
1.28
1.17
0.65
1.15
1.22
1.22
1.08
1.25
1.17
1.14
0.83
1.28
1.1
1.11
0.96
1.18
1.14
1.11
0.99
61.5
67.0
75.3
73.9
87.1
85.2
77.9
69.1
72.4
77.8
58.8
53.2
56.6
51.7
53.1
39.4
85.8
54.0
38.5
62.0
72.5
69.9
39.9
53.7
62.8
66.5
45.8
16.1
69.8
35.2
24.5
15.6
16.3
9.6
9.5
10.4
13.7
21.1
15.7
21.6
26.9
38.6
30.6
23.9
27.9
10.2
22.1
35.9
22.1
21.2
16.8
34.6
27.8
27.9
16.3
27.7
37.4
23.9
3.3
8.5
9.1
9.8
3.3
5.4
11.7
17.2
6.5
6.5
19.5
19.9
4.9
17.7
23.0
32.7
4.0
23.9
25.6
15.9
6.3
13.2
25.5
18.5
9.3
17.2
26.5
46.5
6.3
-------
PI
PI
PI
PI
PI
PI
PI
11
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
Tl
10
10
10
10
10
10
10
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
3
3
3
3
4
A
H
H
H
H
H
H
H
H
B
H
B
B
B
B
B
B
B
B
B
B
B
B
H
B
B
B
B
B
B
am
BID
sm
dm
dm
dm
^^-^-«
&tR
wm
wm
wm
wm
sm
sm
sin
sm
wm
wm
wm
wm
sm
sm
sm
sm
sm
sm
sm
sm
wm
wxn
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
2.1
1.5
1.2
8.1
1.7
1.7
1.2
4.9
2.4
1.8
1.6
6.7
3
2.7
2.3
6
2.5
2
1.7
4.1
2.6
1.7
1.5
8.9
6
3.1
2.6
6.4
2.8
2.9
2.4
1.8
4.1
1.8
2.4
^^•4^M^^«*>-M^«IIV^^— ^V
1.8
5.9
3.5
2.9
1.8
8.8
4.1
2.9
2.4
4.7
2.4
1.8
2.4
7.6
6.5
5.3
4.1
12.9
8.8
4.1
3.5
10.6
4.1
2.2
2.6
2.2
2.2
2.2
2.9
^^^^MI^HM^^—^^M^^
2.6
4.8
2.9
2.2
1.5
3.7
2.9
2.2
1.8
3.7
2.2
2.2
2.2
2.9
2.6
1.8
1.8
20.2
9.6
4.4
3.7
4.0
1.8
7.9
8
7.9
7
7.2
7.7
M^H^q^HMH^H^^^_^^H_
7.8
7.7
7.8
a
8.1
7.3
7.6
7.8
7.9
7.6
7.6
7.7
7.9
7.6
7.7
7.8
7.8
7.3
7.3
7.5
7.6
7.7
7.9
1850
2000
2400
1300
1120
1090
1130
560
1800
2300
3400
1170
2200
2400
2400
1250
2000
2600
2500
420
790
2300
2500
670
530
560
600
800
1530
1.19
1.05
1.15
0.79
1.29
1.25
1.43
0.78
1.05
1.1
0.97
0.89
1.15
1.14
1.11
0.81
1.16
1.11
1.04
1.03
1.15
1.22
1.13
0.85
0.99
1.2
1.19
0.99
1.19
61.8
51.4
45.7
73.8
79.7
77.9
73.1
62.7
50.1
34.6
18.5
39.2
40.0
36.0
28.8
50.1
63.7
42.9
45.1
45.0
37.3
40.0
41.4
29.3
23.4
26.7
25.6
53.2
64.5
23.7
24.8
30.5
23.6
17.5
13.6
^^^^^^V*^M^^MI»M*^^— •*«
14.9
37.3
24.2
24.0
32.1
42.3
32.2
31.9
37.3
49.9
21.5
27.3
19.1
33.3
31.0
29.1
28.8
50.8
49.4
38.9
39.2
37.2
21.7
14.5
23.8
23.8
2.6
2.7
8.5
^*^^^^^*mm*^^lmmmllm**lmm
11.9
0.0
25.7
41.4
49.3
18.5
27.7
32.1
33.9
0.0
14.7
29.8
35.8
21.6
31.7
30.8
29.8
19.8
27.2
34.4
35.1
9.6
13.8
-------
Tl
Tl
Tl
Tl
Tl
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
4
4
4
4
4
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
H
H
H
H
H
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
H
H
H
H
H
B
H
wm
wm
sm
sm
sm
sm
sm
sm
sm
sm
dm
dm
dm
dm
sm
sm
sm
sm
dm
dm
dm
dm
vim
wm
wm
wm
sm
sm
sm
32.5
52.5
7.5
22.5
32.5
52. 5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
2.3
1.7
7.7
4
3.1
2.3
9.5
7.9
13.6
10.6
8.5
10.1
9.8
5.9
9.8
6.6
6.7
8.4
9
9.2
10.3
8.8
11.7
2.5
1.4
1.1
17.4
3
1.5
4.1
2.9
12.9
5.9
4.7
2.9
12.9
13.5
25.3
22.9
20.6
28.2
27.1
18.2
16.5
12.9
10.6
14.7
15.3
12.9
16.5
14.7
11.2
5.3
3.5
2.4
13.5
4.7
3.5
1.5
1.1
7.7
3.0
2.2
1.8
19.9
22.3
8.8
8.1
23.1
22.1
14.0
14.4
15.4
15.1
13.6
6.2
23.2
21.0
19.2
8.8
6.6
3.7
3.3
4.8
5.9
4.4
5.9
7.8
7.9
7.4
7.6
7.3
7.6
7.3
7.2
7.2
7.2
7
7.1
7.5
7.7
7.6
7.2
7.5
7.3
7.3
7.2
7.2
7.4
7
7
7.2
7.2
7.1
7.3
7.2
2100
2900
470
1410
2500
2700
600
1150
1070
1000
820
1020
1190
1100
1650
1700
1600
1450
690
1040
920
800
370
320
250
690
530
500
700
1.05
1.1
0.88
1.02
1.11
1.05
0.79
0.9
0.71
0.79
0.86
0.84
0.86
0.93
0.76
0.96
0.86
0.89
0.82
0.93
0.89
0.89
0.7
1.03
1.32
1.33
0.66
1.21
1.33
64.7
53.5
21.8
18.8
18. S
20.6
16.5
13.7
27.5
24.5
15.3
14.9
20.1
6.0
20.0
28.3
30.8
53.8
22.4
18.5
20.8
44.3
49.5
62.1
79.5
67.7
54.1
72.2
72.9
16.9
22.3
47.8
38.1
41.3
36.3
68.1
70.8
65.5
70.4
68.0
71.5
69.6
66.6
71.7
58.2
56.3
43.6
66.0
73.0
68.9
49.2
44.2
25.6
13.4
18.8
45.8
20.7
18.5
18.5
24.3
30.4
43.1
40.2
43.1
15.4
15.4
7.0
5.1
16.8
13.6
10.3
27.4
8.2
13.5
12.9
2.5
11.6
8.5
10.3
6.5
6.3
12.3
7.1
13.5
0.2
7.2
8.6
-------
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
sm
Cull
uXtt
dm
dm
wm
wm
wm
vim
am
sm
sm
sm
dm
dm
dm
dm
wm
wm
wm
wm
sm
sm
sm
sm
dm
dm
QJH
elm
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
•^-^••••VH^H^^MMVMm
32.5
52.5
7.5
22.5
32.5
52.5
1.3
11.2
5.6
3.1
2.3
6.3
2
1.9
1.2
22.5
6.4
4.8
3.5
19.2
10.9
7.9
3.7
6.9
2.1
1.5
1.5
10.4
3.3
^^^•••IBM-PVtfMtfl^Hla^HAWfl
1.6
1.6
12.2
4.1
3.6
2
2.9
16.5
8.8
7.1
7.1
7.6
3.5
4.1
2.9
11.2
5.9
4.7
3.5
14.7
15.3
10.0
5.3
10.6
4.7
2.9
2.9
11.2
4.7
2.9
2.9
10.0
4.7
5.3
5.9
5.5
24.1
6.6
5.9
7.4
4.4
2.2
2.2
1.8
2.9
3.7
4.0
4.4
4.0
4.4
7.7
10.7
4.4
2.2
1.8
1.8
4.4
3.7
2.6
2.2
3.3
3.7
3.3
2.9
7.3
7.2
7.4
7.8
7.7
7.3
7.7
7.6
7.8
6.9
7.6
7.6
7.7
6.7
6.8
7
7.2
7.8
7.9
7.8
7.7
7.5
7.8
kBH^_MMHHMB^_^v^^^
7.8
7.7
6.9
7.4
7.6
7.6
870
470
500
650
670
310
260
250
250
440
670
760
660
840
1050
950
710
440
320
500
780
610
1150
1080
800
550
500
600
710
1.29
0.7
0.94
1.18
1.18
0.91
1.07
1.09
1.17
0.54
1
1.03
1.09
0.58
0.76
0.93
1.07
0.8
1.13
1.1
1.15
0.84
1.13
•V^^^^^H«^^-V-*VP-*M^^^
1.29
1.22
0.76
1.12
1.07
1.21
66.9
59.9
50.1
55.9
64.5
68.4
66.4
71.1
74.5
40.1
52.9
50.1
47.4
36.0
27.4
24.2
36.2
58.4
64.3
64.4
61.7
56.2
58.2
69.7
66.4
48.7
57.0
51.6
46.1
21.2
38.1
42.7
32.9
24.9
31.6
25.6
23.6
19.5
55.9
40.6
40.1
37.5
60.7
67.2
61.3
41.9
33.6
24.4
21.8
25.0
39.1
28.5
MM>*MMMAM»^^^^^^_^^^ta
18.6
19.0
45.6
32.0
32.1
33.1
11.9
2.0
7.2
11.1
10.5
0.0
8.0
5.3
6.0
3.9
6.5
9.8
15.1
3.3
5.4
14.5
21.9
8.0
11.4
13.8
13.3
4.6
13.3
11.7
14.6
5.8
11.0
16.2
20.8
-------
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
S
6
6
6
6
6
6
6
6
6
6
6
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
3
L
L
L
L
L
L
L
L
!•
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
wm
wm
wm
wm
am
sm
am
am
Cull
dm
dm
CuEl
wm
wm
wm
wm
am
am
am
am
wm
wm
wm
wm
sm
sm
sm
am
wm
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
7.9
2.9
2.5
1.9
13.7
7.8
4.7
3.6
17.8
9.7
7.3
3.9
9.9
10.3
4.3
2.4
10.4
6.6
3.5
7.6
3.5
3.5
2.4
10.6
6.5
4.1
3.5
7.1
5.9
5.3
5.3
12.9
17.6
11.2
5.9
12.4
7.1
3.5
2.2 2.9
2.2 1.8
2.4 1.8
4.5 1 2.9
10 1 7.1
8.8 | 8.2
4.1 4.1
2.2 2.9
1.9 1 2.4
9.4 1 5.9
5.2
2.9
2.9
2.6
2.9
3.3
4.8
6.3
9.2
10.7
9.6
9.5
20.6
11.8
22.5
33.1
26.8
11.8
30.1
48.4
6.6
8.1
8.5
9.2
21.8
22.8
34.6
32.8
12.2
7.5
7.7
7.6
7.7
6.9
7.4
7.2
7.2
6.8
7.1
7.4
7.3
5.4
5.3
5.5
5.7
5.2
5.5
5.5
5.5
rmmrrn ••^^^^^••^rt*^^^
5.7
5.7
5.3
5.2
5.5
5.6
5.7
5.8
5.4
430
300
350
490
600
740
850
690
590
630
590
600
340
350
340
440
300
240
230
400
300
230
170
220
230
200
270
280
230
0.88
1.17
1.21
1.18
0.73
0.91
1.1
1.1
0.61
0.869
0.965
1.08
0.846
0.776
1.09
1.19
0.821
0.99
1.17
1.18
1.23
1.23
1.09
0.84
0.84
1.06
1.22
1.22
0.75
44.3
56.3
48.7
50.8
35.5
31.3
26.0
26.4
25.5
21.3
28.1
17.5
19.6
31.0
29.3
27.9
26.6
30.8
33.0
29.1
32.0
40.3
45.6
45.6
42.5
43.6
40.5
37.2
36.3
55.7
30.1
33.1
26.9
55.4
55.8
50.5
44.8
67.5
66.9
57.6
47.6
68.7
60.5
44.5
37.9
64.6
58.8
43.4
38.6
31.1
33.5
41.4
49.9
57.5
41.3
30.5
30.5
56.6
o.o I
13.7
18.2
22.3
9.1
12 . 9 (I
23.5
28.8 ||
6.9 J|
11.8 ||
14.3 ||
34.8 ||
11.7 ||
8.4 |
26.2 J|
34.2 ||
8.8 ||
10.5 1
23.6 ||
32.3
36.8
26.2
13 . 0
4.5
0.0
15.1
28.9
32.3
7.1 ||
-------
ro
O)
GO
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
C7
3
3
3
3
3
3
3
4
4
4
4
4
4
4
*
L
L
L
L
L
L
L
L
L
L
L
L
L
L
L
wm
wm
wm
am
am
am
am
wm
wm
wm
wm
am
am
am
am
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
7.5
22.5
32.5
52.5
5.6
2.7
2
10
5.8
2.6
1.9
9.6
7.8
4.8
2.7
11
5.6
3.3
1.9
4.1
2.9
1.8
7.1
4.1
1.8
1.8
10.6
6.5
6.5
2.9
8.2
4.1
2.4
1.8
8.5
11.8
35.3
13.6
16.2
52.6
52.9
20.7
11.4
12.5
19.8
30.6
17.6
39.4
34.6
5.4
5.7
5.8
5.2
5.4
5.6
5.7
6
5.8
5.8
6.1
5.3
5.4
5.7
6
200
160
190
170
170
140
240
330
250
300
300
290
190
240
290
1
1.17
1.18
0.78
1.04
1.13
1.18
0.91
0.98
1.03
1.18
0.8
1.05
1.11
1.22
35.8
41.7
45.9
32.9
36.6
40.0
39.4
21.2
29.1
31.2
30.1
33.2
33.5
34.1
27.0
51.1
43.2
34.4
55.3
47.5
44.9
36.2
65.6
60.4
47.7
43.4
57.5
48.6
42.1
39.6
13.1
15.1
19.7
11.8
15.8
15.1
24.4
13.2
10.5
21.1
26.4
9.2
17.9
23.8
33.4
-------
Appendix 7-3
Freeland) and EM-38 data from the CWLSA 1992
I, Trans=transect, V=vertical, H=Horizontal,
„.
WL
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
PI
Tl
Tl
•*^l • • •• I • W%AB
-g
Trans
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
6
6
6
7
7
7
8
8
8
9
9
10
10
10
1
1
"~
Zone
WM
SM
DM
WM
SM
DM
WM
SM
DM
WM
SM
DM
WM
SM
DM
WM
SM
DM
WM
SM
DM
WM
SM
Dm
WM
DM
WM
SM
DM
WM
SM
.
El
V
170
185
130
83
70
82
70
80
67
59
72
64
120
83
105
85
96
96
94
100
100
100
110
110
110
90
115
110
90
165
145
======
«-38
H
125
170
125
60
56
78
48
68
55
38
48
64
90
58
100
64
70
94
70
90
88
88
75
80
90
82
95
85
70
130
110
===:^r==:^===r=^==r
Soil Classification
Typic Fluvaquent
Mollic Fluvaquent
Cum. (Calc) Endoaquoll
Mollic Fluvaquent
Fluvaquent ic Endoaquoll
Cum. (Calc)
Endoaquoll
Aerie Fluvaquent
Typic Calciaquoll
Cum. (Calc) Endoaquoll
Aerie Fluvaquent
Aerie Fluvaquent
Typic Calciaquoll
Aerie Fluvaquent
Typic Calciaquoll
Typic Calciaquoll
Aerie Fluvaquent
Typic Fluvaquent
Typic Calciaquoll
Aerie Fluvaquent
Typic (Calc) Endoaquoll
Typic (Calc) Endoaquoll
Typic Calciaquoll
Aerie Calciaquoll
Typic Calciaquoll
Aerie Fluvaquent
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Calciaquoll
Aerie Calciaquoll
Aerie Calciaquoll
269
-------
Tl
Tl
Tl
Tl
Tl
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
P7
C7
C7
C7
C7
C7
C7
C7
C7
2
2
3
4
4
1
1
2
2
3
3
3
4
4
4
5
5
5
6
6
6
1
1
2
2
3
3
4
4
WH
SM
SM only
WM
SM
SM (no WM)
DM
SM (no WM)
DM
WM
SM
DM
WM
SM
DM
WM
SM
DM
WM
SM
DM
WM
SM
WM
SM
WM
SM
WM
SM
145
135
82
120
nd
70
81
67
68
37
44
46
34
62
74
50
58
54
48
62
72
63
64
64
54
50
54
58
64
115
100
56
85
nd
56
69
56
50
29
32
32
23
45
62
38
40
44
32
44
55
44
50
50
40
36
40
42
46
Aerie Calciaquoll
Typio Calciaquoll
Cumulic (Calc) Endoaquoll
Typic Calciaquoll
Cumulic (Calc) Bndoaquoll
Cumulic Epiaquoll
Cumulic Epiaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Typic Calciaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Aerie Calciaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Typic Calciaquoll
Typic Calciaquoll
Typic Endoaquoll
Typic Calciaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Bndoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
Cumulic Endoaquoll
270
-------
Appendix 9-2-1
Locations of wetlands used for chemistry, sediment,
and/or hormone analysis, 1993.
COUNTY SITE TOWNSHIP RANGE SECTION QUARTER
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
KIDDER
KIDDER
KIDDER
KIDDER
KIDDER
STUTSMAN
STUTSMAN
STUTSMAN
KIDDER
KIDDER
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
KIDDER
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
STUTSMAN
1
2
3
6
7
8
11
13
14
15
16
17
18
20
21
22
23
24
25
26
27
28
29
30
31
39
44
67
77
80
84
87
90
92
96
100
103
115
127
128
129
130
131
144N
144N
144N
14 3K
143N
143N
142N
142N
142N
14 IN
14 IN
14 IN
14 ON
139N
139N
144N
144N
142N
142N
14 IN
139N
138N
139N
144N
14 ON
144N
143N
14 IN
14 ON
140N
139N
139N
139N
13 9N
13 9N
138N
137N
142N
13 9N
142N
142N
139N
139N
69W
69W
69W
63W
63W
63W
68W
66W
66W
69W
69W
68W
69W
68W
67W
71W
71W
71W
71W
72W
69W
67W
67W
72W
70W
68W
63W
66W
67W
67W
69W
68W
66W
66W
65W
66W
66W
71W
69W
66W
68W
69W
69W
16
29
29
21
22
27
01
32
32
06
06
09
22
02
05
20
30
31
31
12
03
04
03
15
34
27
04
04
01
07
07
24
19
12
34
07
34
26
5
32
02
24
24
NW
SE1/4SE1/4
SW1/4SE1/4
SW
SE
NE
SE1/4SW1/4
NE
SW1/4SW1/4
NW
SE
NW
NW
NW
SE
NE
SE
NW
SW
NW
SW
SE
SW
SE
SW
El/2
SW
El/2
Nl/2
Sl/2
ALL
ALL
NE
SE
Nl/2
SW
SW
NE
SW1/4
NW1/4
Sl/2
271
-------
APPENDIX 9-2-2
HORMONAL RESPONSE OF AMPHIBIANS TO
ENVIRONMENTAL STRESS
Steve Dominguez and
Anne Fairbrother
U.S. Environmental Protection Agency
Environmental Research Laboratory
Corvallis, Oregon
Diane L. Larson
U.S. Geological Survey
Northern Prairie Science Center
Jamestown, North Dakota
August, 1993
INTRODUCTION
Harlow et al. (1987) have defined an animal in "stress" as one that is "required to make
abnormal or extreme adjustments in its physiology or behavior to cope with adverse aspects of its
environment". Stress in vertebrates is accompanied by an increase in plasma corticosteroid
concentration (Harlow et al., 1987; Kirkpatrick et al., 1979; Licht etal., 1983; McDonald et a/.,1988;
McDonald and Taitt, 1982; Moore and Deviche, 1987; Moore and Miller, 1984; Orchinik et al., 1988;
Seal and Hoskinson, 1978; Whatley et al., 1977; Wingfield et al., 1982). Often, such increases
accompany a decline in immune system responses which may make stressed populations more
susceptible to disease (Geller and Christian, 1982) and parasitism. Heart rate in domestic sheep is
positively correlated with corticosterone levels (Harlow etal., 1987), suggesting a generally higher cost
of metabolism under stress. High levels of corticosterone have also been associated with decreased or
abolished reproductive behavior in amphibians (Dupont etal., 1979; Moore,1983; Moore and Deviche,
1987).
Parsons (1990) pointed out that the impact of individual environmental stressors cannot be
considered in isolation. Stressors such as environmental contaminants are often difficult and expensive
to measure and their potential synergisms are largely unknown. Because corticosterone release is a
common response to a range of stressors across a wide variety of taxa, measures of plasma
corticosterone concentrations may provide an index of which populations are being stressed. Because
272
-------
. the response is nonspecific relative to the stressor, no assumptions are necessary regarding the
etiology of the stress.
Corticosterone levels may respond to environmental stress in two ways. Baseline levels may
become persistently elevated. This response has been observed in the reptiles Lacerta vivipara
(Duaphin-Villemant and Xavier, 1987) and Urosaurus ornatus (Moore et at., 1991). However, recent
work on birds (J.C. Wingfield, pers. comm.) and amphibians (F.L.Moore, pers. comm.) has indicated
that chronic (e.g., environmental) stress may also affect the rate at which circulating levels of
corticosterone respond to a superimposed acute stress such as handling during capture.
RELATIONSHIP TO ERA'S MISSION
The EPA recently initiated the Environmental Monitoring and Assessment Program (EMAP)-
Wetlands. The program is designed to provide quantitative assessments of the current status and long-
term trends in the ecological condition of wetland resources on both regional and national scales.
EMAP-Wetlands will develop standardized protocols to measure and describe wetlands condition, report
estimates of wetland condition in selected regions across the country, and develop formats for reporting
program results. Longer term goals include trend detection and diagnostic analyses, to identify
plausible causes for degraded or improved wetland condition.
It is proposed that amphibians, both as individuals and in their aggregate populations, would be
good indicators of wetland condition. The life cycle of many amphibians is such that they are
dependent for at least a portion of their life on a wetland habitat. Their relatively low mobility assures
that any stresses they reflect are localized in the sample area. Although population measures and
counts of individuals would provide a crude index of a wetland integrity, it would be preferable to find a
-more sensitive measure of stress such that mitigative changes can be instituted prior to the demise of a
population. Additionally, little is known about natural long-term (10 to 20 year) cycles of amphibians
that potentially could confound a monitoring effort based solely on counts of individuals or population
distributions. It is hoped that biomarkers such as plasma corticosterone concentration can provide the
needed early warning indicator of environmental stress.
273
-------
GOAL
Determine the effect of acute and chronic stress on plasma corticosterone concentrations in a
laboratory population of the tiger salamander, Ambystoma tigrinum, and to test for interactions between
chronic stress and acute stress and time on corticosterone levels.
OBJECTIVES
1. Determine whether underlying chronic stress affects the corticosterone levels achieved
in response to acute handling stress.
H0: The effect of acute handling stress on corticosterone levels is the same in animals
with or without a simultaneous chronic exposure to azinphos-methyl (AZM).
2. Determine the effect of chronic exposure to AZM on corticosterone levels.
3. Determine the effect of acute handling stress on corticosterone levels.
4. Determine whether tail-bleeding causes an immediate detectable acute corticosterone
release.
H0 (2-4): Corticosterone levels are the same in stressed and unstressed animals.
APPROACH DESIGN
Two experiments will be undertaken. Experiment A will involve 144 animals in 36 aquaria. The
experimental unit will be the aquarium. Treatments will be as follows, with animals housed in groups of
four in the 5-gal aquaria:
1. No stress (control) - 48 animals in 12 aquaria.
2. Acute stress only, consisting of 30 minutes of confinement in a 500 ml glass jar half
filled with water just prior to sampling - 32 animals in 8 aquaria.
274
-------
3. Chronic stress (10 days, 20 days) only, consisting of a sublethal concentration (to be
determined by preliminary testing) of azinphos methyl in the water supply - 32 animals
in eight aquaria.
4. Chronic stress as above, with the addition of acute stress as above - 32 animals in
eight aquaria.
This scheme requires two types of water supply - one with AZM directed to 18 aquaria, and
one without AZM directed to the other 18 aquaria. The 36 water lines will be randomized as to which
aquaria they supply.
Experiment B will involve 48 animals in groups of four in 12 5-gal aquaria. The experimental
unit will be the aquarium. Treatments will be as follows:
1. No stress (control) -12 animals in three aquaria.
2. No stress except 90 seconds of simulated tail bleeding just prior to sampling -12
animals in three aquaria.
3. Thirty minutes of confinement in a 500 ml glass jar half filled with water just prior to
sampling -12 animals in three aquaria.
4. Thirty minutes of confinement in a 500 ml glass jar half filled with water, followed by 90
seconds of simulated tail bleeding just prior to sampling -12 animals in three aquaria.
TEST CONDITIONS
Animals trapped in wetland ponds in North Dakota will be housed in flow-through aquaria in
Building P600 at Willamette Research Station (WRS). The water supply will be WRS wellwater. They
will be fed ad libitum with goldfish, crickets, and worms. A light cycle of 13L11D will be maintained.
Testing will be at 20° C, and will be completed before the animals metamorphose out of the aquatic
larval form.
275
-------
Water volume in test aquaria well be set at approximately 10 I by means of a screened
standpipe acting as an overflow drain. Water will flow through the aquaria at 130 ml/min, making the
volumetric turnover time about 77 min. The water supply for aquaria receiving AZM will first pass
through a headbox, where its temperature will be adjusted by a thermostatically controlled heater, and
recorded by a thermograph. It will then flow to a mixing chamber, where AZM in a dimethylformamide
carrier will be injected by a syringe pump, thence to a splitter box and finally to the aquaria. Water
supplies for aquaria not receiving AZM will be tapped directly off the headbox. Aquaria will be
vacuumed clean with a siphon every day.
Once AZM exposure begins, water samples will be taken from two randomly selected aquaria
receiving AZM and two randomly selected AZM-free aquaria each day for confirmation of AZM
concentrations. Relatively few aquaria need to be sampled each day because the system design
assures uniform AZM concentrations across their water supplies, and water turnover rate in the aquaria
is so high that AZM degradation should not be a significant factor, even if flows diverge substantially
from normal for extended periods. Pesticide-free 150 ml glass milk dilution bottles will be filled with
water withdrawn from a depth of approximately 5 cm by means of a 30 ml Manostat pipet. AZM
extraction will normally be done the same day at ERL-C, but can be delayed up to 72 hr post-collection
if samples are refrigerated at 4° C. Extracts will normally be analyzed within 24 hr., but can be held at
4° C for up to 26 days prior to analysis per the Wildlife Ecology Program (WEP) SOP for AZM
analysis.
Water hardness, pH, alkalinity, and conductivity will be measured at least once during the test.
At the beginning of a 7-day acclimation period, designated day -7, animals for Experiment A will
be assigned to the aquaria by stratified random distribution. Each aquarium, in random order, will
receive one animal, then a second set of animals will be distributed, and so on until there are four
animals in each one. The order of placement will be re-randomized for each set. The treatment for
each aquarium, as well as the day the animals therein will be sacrificed and sampled, will already have
been randomly assigned by this time. The first day after acclimation will be designated day 1. On this
day, all animals from four aquaria in Treatment Group 1 (control) will be sacrificed to establish starting
plasma corticosterone levels, and AZM exposure will begin for treatment groups 3 and 4. On days 10
and 20, all animals from four aquaria in each of the four treatments (64 animals each day) will be
sacrificed to complete the test.
276
-------
Experiment B will overlap with Experiment A, utilizing AZM-free aquaria vacated on days 0 and
10 of Experiment A. On Experiment A day 11, Experiment B animals will be assigned to the 12
vacated aquaria by stratified random distribution. After seven days of acclimation, these animals will be
subjected to the Experiment B treatments outlined above.
BLOOD SAMPLE COLLECTION
Blood sampling will be undertaken prior to morning husbandry and feeding, in case those
activities engender transitory corticosterone release. Just prior to sacrifice, an animal will be weighed.
This should take about 15 seconds. It will then be decapitated with scissors, and blood will be
collected from the severed truncus arteriosus in heparinized 70 uJ capillary tubes. As much blood will
be collected as possible. Sets of filled capillary tubes will be plugged with Critoseal and stored in
labeled test tubes on ice until they can be further processed. The severed heads will be held on ice in
labeled plastic bags until they can be frozen at -70° C pending brain cholinesterase analysis.
BLOOD ANALYSIS
After transportation to ERL-C Lab 126, the contents of each set of 70 uJ capillary tubes will be
consolidated into labeled 1.5 ml microcentrifuge tubes, and placed on ice. Each sample will be mixed
with a pipet and enough will be withdrawn to make three smears for lymphocyte differential analysis.
The remainder will be centrifuged in the Son/all RC5C centrifuge for 10 min at 2500 rpm and 4° C. The
plasma layer will then be transferred to a new set of labelled microcentrifuge tubes. If samples are of
sufficient volume, they will be split to allow for plasma cholinesterase analysis. All will be frozen at -70°
C pending analysis.
Hematology and cholinesterase analysis will be per WEP SOPs. Corticosterone analysis will be
performed by Dr. Al Fivizzani at the University of North Dakota.
DATA ANALYSIS
Analysis of variance will be the main approach if data conform to the assumptions of the
technique, or can be transformed to do so. Otherwise, non-parametric tests may have to be used.
Plasma corticosterone concentrations will comprise the key data set. Brain cholinesterase activity may
provide an additional measure of AZM sublethal effect. Chronic elevation of plasma corticosterone
277
-------
concentrations, should it occur, may cause suppression of lymphocyte populations in peripheral blood.
This should be detectable by the lymphocyte differential counts.
ANIMAL WELFARE
A review of this proposal by the ERL-C Animal Care and Use Committee will occur prior to the
initiation of the study. Test animals will be treated in accordance with applicable procedures contained
in "Guidelines for use of Live Amphibians and Reptiles in Field Research" (Am. Soc. Ichthyologists and
Herpetologists, et. al, 1987).
QUALITY ASSURANCE
The data from this study will be used to determine the feasibility of using a plasma
corticosterone biomarker as a monitoring tool for EMAP. Therefore, the data must be of high quality as
they are likely to provide baseline values to which additional field-collected samples can be compared.
A quality assurance project plan will be prepared following ERL-Corvallis guidelines and approved prior
to beginning data collection. Animal rearing, blood collection, handling storage, and analyses will follow
standard operating procedures described in the WEP Quality Assurance Document.
BUDGET
Laboratory supplies and food 500
Cholinesterase reagents 200
Corticosterone analysis (postage) 20
$720
PERSONNEL
Steve Dominguez (Co-Principal Investigator) - study design and laboratory operations, data analysis,
manuscript preparation.
Anne Fairbrother (Co-Principal Investigator) - scientific and technical guidance
Al Nebeker (Co-Investigator) - laboratory operations, data analysis
Diane Larson (Co-Investigator) - study design
Bill Griffis - analytical chemistry (AZM)
278
-------
Tamotsu Shiroyama - biochemistry (cholinesterase)
Lisa Ganio (METI) - biostatistics
TBD (METI) - lymphocyte differential counts
TIMEFRAME
Test animals will be captured in wetland ponds in North Dakota in late July and early August,
1993, and shipped to ERL-C in insulated containers via Federal Express. The study is currently
expected to commence on August 17, with blood sampling on Aug 24 and September 3, 10, and 13.
279
•&U.S. GOVERNMENT PRINTING OFFICE: 1997 - 549-001/«OMO
------- |