xvEPA
Unhed Srami
A LANDSCAPE ATLAS OF ECOLOGICAL VULNERABILITY:
ARKANSAS' WHITE RIVER WATERSHED AND THE
MISSISSIPPI ALLUVIAL VALLEY ECOREGION
Percent Agriculture
B-30
30 - 46
B46-61
61-76
76-92
-------
EPA/600/R-03/057
August 2003
A LANDSCAPE ATLAS OF ECOLOGICAL VULNERABILITY:
ARKANSAS' WHITE RIVER WATERSHED
AND THE
MISSISSIPPI ALLUVIAL VALLEY ECOREGION
^cardo D. Lopez, Daniel T. Heggem, Curtis M. Edmonds, 1K. Bruce Jones,
2Lee A. Bice, 2Matt Hamilton, 2Ed Evanston, ^had L. Cross, and Donald W. Ebert
1U.S. Environmental Protection Agency, Office of Research and Development, National
Exposure Research Laboratory, Environmental Sciences Division, Las Vegas, Nevada
2Lockheed Martin Environmental Services, Las Vegas, Nevada
-------
NOTICE
The information in this document has been funded wholly by the United States Environmental
Protection Agency (EPA) under a Regional Applied Research Effort agreement with EPA
Region 6. It has been subjected to the Agency's peer and administrative review and has been
approved for publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation by EPA for use.
COVER IMAGE DESCRIPTION
Selected study results showing mallard duck habitat suitability models in the (a) Lower White
River Region, mallard duck habitat suitability models in the (b) Mississippi Alluvial Valley
Ecoregion, and water quality vulnerability models in the (c) White River Watershed. (Right)
The Lower White River Region mallard duck winter habitat suitability model with an overlay
map of the mallard duck habitat Unified Vulnerability Index for the South Unit of the White
River National Wildlife Refuge. (Lower left) Hillshaded digital elevation model of the White
River Watershed, overlaid with a water quality vulnerability model based on percent agriculture.
-------
TABLE OF CONTENTS
(CLICK ON LINK)
Cover
Title Page
Notice and Cover Image Description
List of Tables
List of Figures
List of Appendices
Executive Summary
Acknowledgements and Dedication
Chapter 1. Landscape Analysis Overview
Project Overview and Summary of Regional Environmental Issues
Objectives
Approach and Methods
Habitat Vulnerability Assessments
Water Quality Vulnerability Assessments
The Landscape-Ecological Perspective
Data Reporting and Statistics
Chapter 2. Habitat Assessments
Study Areas
How to Navigate and Interpret Habitat Assessment Maps
Habitat Suitability
Mallard Duck Winter Habitat
Black Bear Wetland Habitat
Least Tern Breeding Habitat
Wetland Plant Habitat
Habitat Suitability Maps
Habitat Vulnerability
-------
Chapter 3. Water Quality Vulnerability Assessments
Study Area
How to Navigate and Interpret Water Quality Vulnerability Maps
Water Quality Vulnerability
Roads and Agriculture
Natural Vegetation
Chapter 4. Assessment of Habitat Vulnerability in Wildlife Refuges
National Wildlife Refuges in the Lower White River Region
Results
Discussion
Recommendations and Conclusion
Simulating Landscape Change in the White River National Wildlife Refuge
Acronyms and Abbreviations Used
Appendices
Literature Cited
-------
LIST OF TABLES
(CLICK ON LINK)
Table la. GIS data sets used to derive mallard duck, black bear, least tern, and wetland plant
habitat suitability and habitat vulnerability in the Lower White River Region.
Table Ib. GIS data sets used to derive mallard duck, black bear, least tern, and wetland plant
habitat suitability and habitat vulnerability in the Mississippi Alluvial Valley Ecoregion.
Table Ic. GIS data sets used to derive water quality vulnerability models in the White River
Watershed.
Table 2a. Data classes used to produce habitat suitability GIS models for mallard ducks in the
Lower White River Region and the Mississippi Alluvial Valley Ecoregion.
Table 2b. Data classes used to produce habitat suitability GIS models for black bears in the
Lower White River Region and the Mississippi Alluvial Valley Ecoregion.
Table 2c. Data classes used to produce habitat suitability GIS models for least terns in the
Lower White River Region and the Mississippi Alluvial Valley Ecoregion.
Table 2d. Data classes used to produce habitat suitability GIS models for wetland plants in the
Lower White River Region and the Mississippi Alluvial Valley Ecoregion.
Table 3a. Data classes used to produce habitat vulnerability GIS models for mallard ducks,
black bears, and wetland plants in the Lower White River Region and the Mississippi Alluvial
Valley Ecoregion.
Table 3b. Metrics used to produce water quality vulnerability GIS models in the Lower White
River Region, measured among 8-digit Hydrologic Unit Codes (HUCs).
Table 4. National Land Cover Dataset (NLCD) categories.
-------
Table 5. Spearman Rank Correlation of surface water quality parameters and percent agriculture
land cover within 8-digit HUC subwatersheds in the White River Watershed.
Table 6. Spearman Rank Correlation of surface water quality parameters and forest metrics
within 8-digit HUC subwatersheds in the White River Watershed.
Table 7. Area of potential mallard duck winter habitat in the Lower White River Region.
Table 8. Area and perimeter of each of the 72 federal refuge zones in the Lower White River
Region, in ascending order of refuge zone area.
Table 9. Summary table of Spearman Rank Correlation between federal refuge zone area and
habitat vulnerability parameters.
Table 10. Rank of each of the 72 federal refuge zones in the Lower White River Region by (a)
zone area, (b) zone perimeter, (c) mean area of habitat patches within a zone, and (d) mean
percent contribution of habitat patches within a zone.
Table 11. Rank of each of the 72 federal refuge zones in the Lower White River Region by
habitat patch characteristics.
Table 12. Rank of each of the 72 federal refuge zones in the Lower White River Region by
habitat patch human-induced disturbance characteristics.
Table 13. Assessment of the presence of wetlands less than or equal to 2 ha within each of 68
federal refuge zones (i.e., excluding the four largest refuge zones) in the Lower White River
Region.
-------
LIST OF FIGURES
(CLICK ON LINK)
Figure 1. Geographic overview of the Lower White River Region (LWRR) and White River
Watershed (WRW) study areas.
Figure 2a. Orientation map of Lower White River Region (LWRR), Mississippi Alluvial Valley
Ecoregion (MAVE), and White River Watershed (WRW). Inset shows the states that intersect
the MAVE study area.
Figure 2b. Orientation map of Lower White River Region (LWRR), including the 13 counties
that intersect the LWRR study area.
Figure 2c. County map of all study areas including the 170 counties in Arkansas, Missouri,
Louisiana, Tennessee, Illinois, and Kentucky that intersect the Lower White River Region
(LWRR), Mississippi Alluvial Valley Ecoregion (MAVE), and the White River Watershed
(WRW).
Figure 3. Land cover (National Land Cover Dataset; NLCD) in the White River Watershed,
showing subwatershed boundaries.
Figure 4. Wetland conversion to crop agriculture, pasture, or urban land cover (pre-1600 to
1992) in the Lower Mississippi River basin.
Figure 5. "Loading logs on a lumber company railroad in early 1900" in an Arkansas
bottomland hardwood swamp.
Figure 6. Diesel generator pumping groundwater for the irrigation of row crops in Arkansas
County, Arkansas.
Figure 7. The Beouf-Tensas Project, Arkansas-Louisiana, an example of riparian vegetation
destruction and wetland hydrologic alteration in the Mississippi Alluvial Valley Ecoregion.
-------
Figure 8. The McClellan-Kerr Navigation Channel on the Arkansas-Louisiana border, an
example of riparian vegetation destruction and wetland hydrologic alteration in the Lower White
River Region (LWRR).
Figure 9a. Lower White River Region reference image with selected towns indicated with blue
arrows. Wildlife refuge zones are indicated by white polygons. Single near-infrared band
image.
Figure 9b. Lower White River Region reference image with selected towns indicated with blue
arrows. Wildlife refuge zones are indicated by white polygons. Three band false-color infra-red
composite image.
Figure 9c. Lower White River Region reference image with selected towns indicated with blue
arrows. Wildlife refuge zones are indicated by white polygons. Three band false-color
'enhanced vegetation' infra-red composite image.
Figure 10. Land cover in the White River Watershed study area was measured within
cumulative 30 m riparian "buffer zones", from shorelines to a maximum of 300 m.
Figure 11. How to interpret the habitat suitability, habitat vulnerability, and water quality
vulnerability maps in this atlas.
Figure 12. Hierarchical schematic of the mallard duck winter habitat model process for the
Lower White River Region (LWRR).
Figure 13. An "emergent wetland" in the Lower White River Region.
Figure 14. A "scrub/shrub wetland", across a pool with duckweed (Lemna sp.) and watermeal
(Wolffia sp.) on the surface; Lower White River Region.
Figure 15. A "forested wetland", in the Lower White River Region.
-------
Figure 16. "Unconsolidated bottom wetland" types in the Lower White River Region.
Figure 17. An "aquatic bed" with submersed and emergent vegetation present.
Figure 18. A portion of the White River in the Lower White River Region study area. River
areas provide areas for animals to feed, drink, and rest.
Figure 19. Features and examples of classes and hydrologic modifiers in palustrine wetlands.
Figure 20. Sandy shore of the White River in the Lower White River study area.
Figures 21 a - m. Lower White River Region (LWRR) mallard duck, black bear, least tern, and
wetland plant habitat suitability models.
Figures 22 a - r. Lower White River Region (LWRR) mallard duck, black bear, and wetland
plant habitat vulnerability models (> 2 ha) in terms of habitat patch size and shape.
Figures 23 a - x. Lower White River Region (LWRR) mallard duck, black bear, and wetland
plant habitat vulnerability models (> 2 ha) in terms of human-induced disturbance factors.
Figures 24 a - m. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck, black bear,
least tern, and wetland plant habitat suitability models.
Figures 25 a - r. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck, black bear, and
wetland plant habitat vulnerability models (> 2 ha) in terms of habitat patch size and shape.
Figures 26 a - x. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck, black bear, and
wetland plant habitat vulnerability models (> 2 ha) in terms of human-induced disturbance
factors.
-------
Figure 27. Twenty-five 8-digit hydrologic unit code (HUC) subwatersheds in the White River
Watershed. Yellow triangle indicates a National Water Quality Assessment (NAWQA) Program
field sampling location, sampled during the 1990s (N = 35).
Figure 28. Location of surface water in the White River Watershed. Surface water location data
was used to calculate the percent of streams in proximity to roads.
Figure 29. Road network in the White River Watershed. Data used to calculate the percent of
streams within thirty meters of a road.
Figure 30. White River Watershed percent of streams within 30 meters of a road.
Figure 31. Proposed water quality vulnerability metrics, based on percent agriculture (i.e.,
NLCD codes 81, 82, 83, and 85 in Table 4) adjacent to shorelines, within cumulative thirty-
meter riparian zones, and within entire HUCs.
Figure 32. Proposed water quality vulnerability metrics, based on percent crop agriculture (i.e.,
NLCD codes 82, 83, and 85 in Table 4) adjacent to shorelines, within cumulative thirty-meter
riparian zones, and within entire HUCs.
Figure 33. Proposed water quality vulnerability metrics, based on percent pasture agriculture
(i.e., NLCD codes 81 in Table 4) adjacent to shorelines, within cumulative thirty-meter riparian
zones, and within entire HUCs.
Figure 34. Hillshaded digital elevation model (DEM) of the White River Watershed, depicting
areas from lower elevation (darker) to higher elevation (lighter). This DEM was used to
calculate percent of total agriculture on slopes greater than three percent.
Figure 35. White River Watershed percent total agriculture on slopes greater than three percent.
-------
Figure 36. Spatial analysis of twenty-five 8-digit hydrologic unit code (HUC) subwatersheds in
the White River Watershed. There is a relatively greater risk of surface water quality impairment
in 2 HUCs as a result of the presence of agriculture within the cumulative 120 meter riparian
zones.
Figure 37. Proposed water quality vulnerability metrics, based on (a) largest forest patch
proportion of HUC, (b) mean area of forest patch, (c) largest forest patch area, (d) forest patch
density, (e) forest patch number, and (f) percent of HUC that is forest (by area).
Figure 38. Proposed water quality vulnerability metrics, based on percent forest (i.e., NLCD
codes 41, 42, and 43 in Table 4) adjacent to shorelines, within cumulative thirty-meter riparian
zones, and within entire HUCs.
Figure 39. Proposed water quality vulnerability metrics, based on percent wetland (i.e., NLCD
codes 91 and 92 in Table 4) adjacent to shorelines, within cumulative thirty-meter riparian zones,
and within entire HUCs.
Figure 40. Proposed water quality vulnerability metrics, based on percent natural land cover
(i.e., NLCD codes 31, 41, 42, 43, 51, 71, 91, and 92 in Table 4) adjacent to shorelines, within
cumulative thirty-meter riparian zones, and within entire HUCs.
Figure 41. Spatial analysis of twenty-five 8-digit hydrologic unit code (HUC) subwatersheds in
the White River Watershed. There is a relatively greater risk of surface water quality impairment
in 6 HUCs as a result of forest loss in riparian zones of various widths.
Figure 42. Seventy-two federal refuge zones in the Lower White River Region were compared
using mallard duck winter habitat vulnerability models.
-------
Figure 43. The unified vulnerability index was applied to a hypothetical landscape where
change in the riparian wetland hydrology has occurred. Red rectangle indicates the portion of
the South Unit in the White River National Wildlife Refuge where landscape change is
hypothesized.
Figure 44. (A) Mallard duck winter habitat vulnerability under current hydrologic conditions in
the South Unit of the White River National Wildlife Refuge; (B) Loss of mallard duck winter
habitat under hypothetical decrease in flood stage and duration on the White River; (C) enlarged
view of predicted net loss of mallard duck winter habitat under hypothetical decrease in flood
stage and duration on the White River.
-------
List of Appendices
(Click on Link)
Appendix A. U.S. Fish and Wildlife Service Publication: Habitat Suitability Index Models:
Mallard (Winter Habitat, Lower Mississippi Valley)
Appendix B. U.S. Fish and Wildlife Service Publication: Habitat Suitability Index Models:
Black Bear, Upper Great Lakes Region
Appendix C. U.S. Fish and Wildlife Service Publication: Habitat Suitability Index Models:
Least Tern
Appendix D. U.S. Fish and Wildlife Service Publication: Classification of Wetlands and
Deepwater Habitats of the United States
-------
ACKNOWLEDGEMENTS
We thank Gerald Carney, Jack Hill, Richard Hines, Barbara Keeler, Joe Krystofik, Larry Mallard, Ryan Mollnow,
Sharon Osowski, Betsy Smith, Kenneth Teague, and Donald Williams for their comments regarding this manuscript.
Special thanks to Barbara Keeler and Norman E. Dyer of the U.S. Environmental Protection Agency's Region 6
Office in Dallas, Texas for their support of this project. The authors thank the entire staff of the White River
National Wildlife Refuge, Cache River National Wildlife Refuge, and the Bald Knob National Wildlife Refuge for
their ongoing support of this research.
DEDICATION
From Mark Twain's, Life on the Mississippi - 1883
Now when I had mastered the language of this water and had come to know every trifling
feature that bordered the great river as familiarly as I knew the letters of the alphabet, I had
made a valuable acquisition. But I had lost something, too. I had lost something which could
never be restored to me while I lived. All the grace, the beauty, the poetry had gone out of the
majestic river! I still keep in mind a certain wonderful sunset which I witnessed when
steamboating was new to me. A broad expanse of the river was turned to blood; in the middle
distance the red hue brightened into gold, through which a solitary log came floating, black
and conspicuous; in one place a long, slanting mark lay sparkling upon the water; in another
the surface was broken by boiling, tumbling rings, that were as many-tinted as an opal; where
the ruddy flush was faintest, was a smooth spot that was covered with graceful circles and
radiating lines, ever so delicately traced; the shore on our left was densely wooded, and the
somber shadow that fell from this forest was broken in one place by a long, ruffled trail that
shone like silver; and high above the forest wall a clean-stemmed dead tree waved a single
leafy bough that glowed like aflame in the unobstructed splendor that was flowing from the
sun. There were graceful curves, reflected images, woody heights, soft distances; and over the
whole scene, far and near, the dissolving lights drifted steadily, enriching it, every passing
moment, with new marvels of coloring.
-------
EXECUTIVE SUMMARY
This report presents the potential for loss or damage to
ecosystems in Arkansas' White River Watershed and
the Mississippi Alluvial Valley, where approximately 90%
of wetlands and a substantial percent of other natural
areas have been destroyed. The report is a synopsis of
results to date and subsequent updates to this information,
including other presentations, publications, and data can be
obtained at our web site: http://www.epa.gov/nerlesd1/
land-sci/whiteriver.htm.
This research synthesizes the theoretical foundations of
ecological function with landscape-ecological analyses,
such that the potential interactions between the ecology and
human land use of the region can be depicted. Several
technical approaches are discussed, each of which involve
the use of geographic information systems. The technical
approach in this report interlinks a tremendous amount of
ecological information, which is used to address the question
of ecological vulnerability. Future work includes (a) a
comparative discussion of developing habitat assessment
techniques such as those in the report, (b) further exploration
of selected habitat suitability/vulnerability parameters used in
the study areas, and (c) further exploration of selected water
quality parameters in the White River Watershed.
Chapter 1 discusses the ecological and societal bases of
this research. Chapter 2 describes the approach of using
landscape metrics to assess habitat vulnerability for several
taxa, using patch size, patch shape, road, and human popula-
tion density metrics throughout the entire Lower White River
Region and Mississippi Alluvial Valley Ecoregion. An example
of the mapped results in this report is the unified vulnerability
index, which combines patch size and shape vulnerability metrics
with human-induced disturbance metrics to depict mallard duck
foraging habitat vulnerability in the Lower White River Region (right).
The use of other "landscape" metrics to measure water quality vulner-
ability is discussed and applied to the entire White River Watershed
in Chapter 3. Specific applications of the unified vulnerability index
are discussed in Chapter 4, specifically demonstrating the relative
vulnerability of wildlife refuges and the potential impacts of future
landscape change on habitat vulnerability in riparian areas of the
White River.
The systems approach taken in this report is a substantial step toward
better understanding the systemic answer to the question, "how vulnerable
is the ecology of the White River region?" However, an all-encompassing
answer to the question of ecological vulnerability is multtfaceted, and does
not take the form of a single answer. The problem of determining the single,
most important vulnerability factor within ecological systems is a fundamental
'system-level analysis' problem, not unique to ecology. Despite such analytical
complexities, the authors have created this report in a digital format that best
conveys their results, in consultation with local, regional, and national environ-
mental professionals. Chapter 1 includes several suggested readings for those
who choose to explore the special topics discussed in this report.
Mallard Winter Habitat - Unified Vulnerability Index, patches > 2 hectares
|H < -3 Std Oev. (Greatest Vulnerability)
•H -3 - -2 Std. Dev.
-2 - -1 Std. Dev.
-1 - 0 Std. Dev.
Mean
M 0- 1 Std. Dev.
1-2Std. Dev.
B2 - 3 Std. Dev.
> 3 Std. Dev. (Least Vulnerability)
-------
CHAPTER 1. LANDSCAPE ANALYSIS OVERVIEW
Project Overview and Summary of Regional Environmental Issues
The White River begins in mountainous northwestern Arkansas, flows through
southwestern Missouri, reenters north central Arkansas, and flows down from the Ozark
Mountains into Arkansas' agricultural plain, where it meanders to its confluence with the
Mississippi River (Figure 1). The catchment area of the White River Watershed (WRW; Figure
2a) extends from the Fayetteville, Arkansas in the western Ozark Mountains to the Mississippi
River, and drains from a wide range of landscapes containing farmland, upland forests, wetlands,
lakes, streams, and urban areas (Figure 3). There are seven major dams that maintain large
reservoirs along the White River, and a National Scenic River (Figure 1). Along the banks of the
White River and its tributaries there are two National Wildlife Refuges, two National Forests,
and a National Scenic River. The Cache River (a tributary to the White River; Figure 1) and its
wetlands have been designated as Ramsar Wetlands of International Importance (Ramsar, 2002),
along with 1,235 other wetlands around the world. The "Lower White River Region" (Figure 2a
and Figure 2b) contains most of the White River channel that flows through Arkansas'
agricultural plain, which is a region currently dominated by row-crop agriculture (Figure 3) but
also contains a large proportion of the Mississippi River Valley's last-remaining bottomland
hardwood wetland 'swamps' (Dahl, 1990; Figure 4). Thus, the Lower White River Region
(LWRR) was used in this project, in addition to the entire Mississippi Alluvial Valley Ecoregion
(MAVE), to assess the ecological vulnerability of habitat throughout the region (Figure 2a).
The LWRR is unusual because (a) it provides suitable habitat for the largest winter
concentration of mallard ducks (Anas platyrhynchos) in North America, (b) provides necessary
habitat for recovering populations of black bears (Ursus americanus), (c) provides critical shore
habitat for the endangered populations of least terns (Sterna antillarum), and (d) provides some
of the last-remaining habitat for wetland plants in the region (Carreker, 1985; Allen, 1987;
Rogers and Allen, 1987). The White River aquatic ecosystems also support an important
riverine fishery, including sturgeon (Scaphirhynchus albus and S. platorynchus) and paddlefish
(Polyodon spathula\ and aquatic plant communities within the bottomland hardwood swamps,
which represent some of the most biologically diverse and productive ecosystems of the world
(Mitsch and Gosselink, 1993). The White River and the surrounding landscape also contain
-------
valuable resources for the people living and working in the region because they provide people
with plentiful irrigation water, clean drinking water, flood control, transportation for agricultural
commerce, recreation, commercial shelling and fishing products, and tourism.
Although the predominant land cover in the LWRR and MAVE is currently agriculture,
forest land-cover predominates in the Ozark Mountain region of the WRW (Figure 3). Closer
inspection of the land cover throughout the WRW reveals that the landscape is a mosaic of many
different discernable land cover types, such as dry "upland" forests, wetlands, human-built and
populated areas, pastures, and row-cropped land. This mixture of land cover occurs, for
example, as a gradient (i.e., a gradual change in the relative proportions) of land cover types
from the northwestern corner of the WRW to the southeastern corner of the WRW (Figure 3).
Similar gradients to this are used in this assessment to predict habitat suitability (i.e., the
applicability of land cover to organismal requirements) for mallard ducks, black bears, least
terns, and wetland plants; to predict habitat vulnerability (i.e., the risk of loss or damage of
suitable habitat) for mallard ducks, black bears, least terns, and wetland plants; and to predict the
vulnerability of surface water to impairment from runoff. Vulnerability predictions were
performed with the use of currently available land cover data for the Lower White River Region,
the Mississippi Alluvial Valley Ecoregion (Omernik, 1987), and the White River Watershed
(Figure 2; U.S. GS, 1994).
Historically, the Lower White River Region, the Mississippi Alluvial Valley Ecoregion,
and the White River Watershed have all undergone substantial alterations in land cover (Dahl,
1990; Figure 4), particularly the conversion of wetlands (Figure 5) to agricultural land (Figure 6).
Consequently, there has been tremendous biological and hydrologic change throughout the
landscape of the region (Dahl, 1990; Mitsch and Gosselink, 1993), particularly in riparian
wetlands (Figure 7; Figure 8). Although the majority of wetland losses in the region occurred
prior to the 1970s, the trend has continued in Arkansas, Mississippi, and Louisiana as a result of
wetland conversion (Johnston, 1989; Dahl and Johnson, 1991; The Nature Conservancy, 1992;
Kress et al., 1996; Heggem et al., 2000; NRI, 2000). In particular, seventy-percent of Arkansas'
wetlands have been destroyed since the late nineteenth century (Dahl, 1990), a loss of
approximately 28,000 square kilometers of wetland, with greater than 4,000 square kilometers of
wetland loss occurring in Arkansas during the first half of the twentieth century (Shaw and
Fredine, 1956). The ongoing losses of wetlands in the region have been positively correlated
-------
with regional and local losses of biological diversity (Gosselink and Turner, 1978; Ewel, 1990;
Kilgor and Baker, 1996; Smith, 1996; Wakeley and Roberts, 1996); an increase in frequency,
severity, and duration of flood events (Hopkinson and Day, 1980a; Hopkinson and Day, 1980b;
Brown, 1984); and the degradation of downstream water quality (Kitchens et al., 1975; Day et
al., 1977; Hupp and Morris, 1990; Hupp and Bazemore, 1993; DeLaune et al., 1996; Dortch,
1996; Kleiss, 1996; Long andNestler, 1996; Walton et al., 1996a; Walton et al., 1996b; Wilber
et al., 1996). This study is a first step towards determining how landscape scale (i.e., broad
scale) land-cover changes may have influenced habitat loss or degradation, and surface water
quality. This map-based "atlas" geographically depicts and quantifies some of the
aforementioned relationships, and endeavors to provide information for future land use planning
in the Lower Mississippi River region. This atlas focuses primarily on the loss and degradation
of wetlands because future development in the vicinity of the White River has great potential for
altering wetland and aquatic ecosystem habitat conditions, regional hydrology, and water quality.
The results in this atlas are timely because there are several land development projects and
human activities in the vicinity of the White River that are planned or ongoing, each of which
has the potential to damage or destroy a substantial portion of the remaining wetlands in the area.
In general, planned or ongoing development activities in the area include: construction of dikes
or channel modifications to increase river flow rates (personal communications with U.S. Fish
and Wildlife Service and U.S. Army Corps of Engineers, 2001); agricultural irrigation projects
that involve increasing the removal of surface water from the White River to supplement
agriculture irrigation shortages (Jehl, 2002); modification of reservoir release schedules
(personal communication with U.S. Fish and Wildlife Service, 2000); and road construction
projects (Arkansas State Highway Commission, 2002).
The White River has never before undergone a landscape-scale ecological assessment of
this kind despite the fact that it contains important and rare habitat, is one of the major tributaries
to the Mississippi River, and contributes a large amount of nutrients to the Mississippi River and
the Gulf of Mexico (Presely et al., 1980; Meade, 1996; Rabalais et al., 2002). Thus, an
ecological vulnerability assessment of the White River and surroundings is warranted to
understand better the landscape-scale ecological relationships in the region and to contribute to
the decision-making processes during future land development planning. To this end, this atlas
was completed using readily available geographic information system (GIS) data, provides
-------
several forms of maps, and uses simple model calculations that consist of understandable
components. Thus, the results in this atlas may be easily browsed (in the digital version of the
report) and analyses may be reconstructed, recalculated, or applied to other data sets in the
future.
Objectives
This atlas addresses two project objectives: (1) determining the risks of habitat damage
or loss within the LWRR and the MAVE (i.e., habitat vulnerability) and (2) determining the risks
of water quality impairment within the WRW (i.e., water quality vulnerability). The GIS-based
habitat models are applied to the entire MAVE and to the LWRR using similar techniques, but
with different data sets because the larger MAVE does not have the full coverage of GIS data
that is available for the LWRR (see Approach and Methods). The results depicted in this atlas
further the goals of U.S. EPA's Landscape Sciences Program [e.g., see Jones et al., 2000a] by
specifically contributing to the progressive strategy of (i) detecting cross-sectional landscape
change by comparing thematic data sets in a GIS environment; (ii) quantifying landscape
change; (iii) investigating landscape metrics (e.g., percent forest cover); and (iv) developing
landscape indicators (e.g., forest cover as a indicator of surface water total phosphorus
concentration).
Approach and Methods
Habitat Vulnerability Assessment
Mallard ducks, black bears, least terns, and wetland plants were selected for the GIS-
based habitat vulnerability modeling in this atlas because: (a) sufficient habitat suitability
literature was available for all taxa; (b) GIS data coverages were available and sufficient to
represent published habitat requirements for all taxa; (c) the selected taxa require either wetland
or shoreline conditions during at least a portion of the year; (d) the selected taxa have either
undergone a population decline at some time in the study region, are recovering, or are presently
listed as endangered; and (e) the selected taxa are of special interest to local, regional, or national
natural resource professionals.
-------
The habitat vulnerability assessment objectives of this project were met by addressing
five landscape-ecological hypotheses within the relatively fine-scale LWRR and relatively
broad-scale MAVE (Figure 2). For clarity, the hypotheses are stated below as questions:
Question 1. What are the gradients of land cover in the LWRR (Table la) and the MAVE
(Table lb)?
Question 2. How do the observed gradients of land cover in the LWRR and the MAVE affect
the suitability of habitat for mallard ducks (Table 2a), black bears (Table 2b),
least terns (Sterna antillarum; Table 2c), and regional wetland plants (Reed, 1988;
Table 2d) based on basic organismal requirements?
Question 3. How do the observed gradients of patch size, patch shape, and human-induced
disturbance within the LWRR and the MAVE (Table 3a) affect the vulnerability
of mallard duck, black bear, least tern, and wetland plant habitat?
Question 4. How is mallard duck habitat suitability (from question 2) and mallard duck habitat
vulnerability (from question 3) distributed among the seventy-two refuge areas
that comprise the White River National Wildlife Refuge (NWR), Cache River
NWR, and Bald Knob NWR (Figure 9a)?
Question 5. What mallard habitat losses are likely to occur in the South Unit of the White
River NWR, given current habitat vulnerability and a hypothetical decrease in
flooding of river-adjacent riparian wetlands (Figure 9a)?
Water Quality Vulnerability Assessment
The water quality vulnerability assessment objectives of this project were met by
addressing two landscape-ecological hypotheses within the WRW (Figure 2). For clarity, the
hypotheses are stated below as additional questions:
Question 6. What are the landscape gradients among twenty-five 8-digit hydrologic unit code
(U.S. GS, 1994) 'HUC subwatersheds' (Figure 2a) based on current land cover
(Table lc)?
Question 7. How do the observed landscape gradients of land cover potentially affect surface
water quality in the study areas, based on previously published correlations and
-------
validation with 1990s National Water Quality Assessment (NAWQA) Program
surface water physio-chemical data (U.S. GS, 2001)?
Water quality vulnerability assessments were performed by measuring land cover within a given
8-digit hydrologic unit code (HUC) subwatershed and within cumulative 30m riparian zones to a
maximum of 300 m from shorelines within a HUC. The land-cover-derived metrics (Table Ic
and Table 3b) within each 30m riparian zone (Figure 10) and within each complete HUC were
then compared among HUCs to compare the scale-dependency of such measurements (Figure
11).
The Landscape-Ecological Perspective
This study makes use of ecological indicators, landscape metrics, and landscape
indicators. For the purposes of this atlas an 'ecological indicator' is a sample measurement of an
ecological resource (Bromberg, 1990; Hunsaker and Carpenter, 1990; Hunsaker et al., 1990),
typically from field sampling (e.g., a total phosphorus concentration at a single gauging station
on the Cache River). When measured at a relatively broad 'landscape scale' (Forman, 1995),
'landscape metrics' (e.g., percent forest cover) that are characteristic of the environment, as
measured by a sufficient sample size of ecological indicators can provide quantitative
information about ecological resources at broad scales, and are referred to as 'landscape
indicators' (Jones et al., 1997). In this atlas we selected landscape metrics (Table 3a and Table
3b) that are correlated with ecological indicators at several scales, i.e., at relatively broad scales
(e.g., Riitters et al., 1995; Jones et al., 2000b; Jones et al., 2000c; Jones et al., 2001), at moderate
scales (e.g., van der Valk and Davis, 1980; Roth et al., 1996; Nagasaka and Nakamura, 1999;
Fauth et al., 2000; Lopez et al., 2002; Lopez and Fennessy, 2002), and at single-site or
mesocosm scales (e.g., Peterjohn and Correll, 1984; Murkin and Kadlec, 1986; Ehrenfeld and
Schneider, 1991; Willis and Mitsch 1995; Mclntyre and Wiens, 1999a; Luoto, 2000). The
combined use of these previously observed correlations, GIS mapping techniques, and statistical
analysis techniques across the LWRR, MAVE, and WRW study areas facilitated the
determination of correlations between land cover gradients and ecological vulnerability at each
scale, further aided by relatively rapid computing rates and large computer memory storage
space available on currently available personal computers (Scott et al., 1993; Jones et al., 1997).
-------
This atlas makes use of different landscape gradients. Each gradient is a range of a
condition, which is observed across a selected landscape unit (e.g., across the LWRR) or among
reporting units (e.g., among HUCs or among wildlife refuges). Thus, the two important selection
criteria for study areas in a landscape gradient assessment are a sufficient range of conditions
along each landscape gradient of interest within a study area, and a sufficient number of sites to
compare among reporting units (Green, 1979; Karr and Chu, 1997). Accordingly, this atlas uses
a complete coverage of GIS data (i.e., a "wall-to-wall" coverage of land cover data) to produce
maps in the LWRR, MAVE, and WRW study areas. An initial visual analysis of available GIS
and remote sensing data (Table la, Table Ib, and Table Ic), and meetings with local experts
from the Arkansas Game and Fish Commission, the Arkansas Natural Heritage Commission, the
Arkansas Soil and Water Conservation Commission, The Nature Conservancy, the U.S. Army
Corps of Engineers, the U.S. Environmental Protection Agency; and the U.S. Fish and Wildlife
Service also indicated sufficient land cover variability within each study area to conduct valid
gradient analyses.
There is an inherent trade-off between conducting site-based studies, which are limited
by a lack of contextual information about the surrounding landscape, and landscape-scale
studies, which are limited by a lack of detailed information about small areas. Therefore,
although the landscape-scale assessments in this atlas are founded on the ecological principals of
site-specific studies, the enclosed models may be limited by a lack of detailed information about
small habitat areas. Thus, these models are intended as a preliminary screening tool for large
areas that would otherwise be impractical to assess in the field. For best results, these GIS-based
models should be used in combination with detailed field investigations.
Data Reporting and Statistics
The habitat suitability maps (Figure 11) depict the different habitat requirements for
mallard ducks (Table 2a), black bears (Table 2b), least terns (Table 2c), and wetland plants
(Table 2d). Habitat suitability maps are color-coded to best distinguish between habitat types
and patches. Habitat vulnerability maps depict the relative risk of habitat loss or damage for
each taxon or guild (i.e., wetland plants) as a result of patch size, patch shape, human-induced
disturbance(s), or a combination of these parameters. Habitat vulnerability maps are color-coded
-------
from greatest vulnerability (red tones) to least vulnerability (blue tones), based on standard
deviations from the mean of each vulnerability parameter (Figure 11).
In each water quality vulnerability map the data range in the legend shows the metric
values within each of the 8-digit HUC sub water sheds, contained in each of five quintiles (Figure
11). A quintile contains one-fifth of the distribution of each watershed metric. Quintiles are
formed after ranking watersheds for each metric. The map of the WRW is color-coded to show
relative conditions among watersheds, ranging from red (i.e., greatest vulnerability) to green
(i.e., least vulnerability), potentially indicating a relative range in terms of surface water quality
impairment from nutrient or sediment loading.
Because one or both of the assumptions of parametric statistics tests (normality and
equality of variance) are violated in all of the data, correlation analyses for all parameters were
completed with Spearman Rank Correlation (Zar, 1984), a = 0.05. All statistical analyses were
computed with Statview software (SAS Institute, v.5.0.1, Gary, North Carolina). All GIS
calculations, gradient analyses, and mappings were performed using Arc View GIS software
(Environmental Systems Research Institute, v.3.2, Redlands, California) and ATtlLA GIS
software (U.S. EPA, v.2.999, Las Vegas, Nevada). To allow for visual comparison of relative
habitat vulnerability across the study area landscapes, vulnerability each metric was displayed as
a mean value and standard deviations from the mean (Figure 11).
-------
CHAPTER 2. HABITAT ASSESSMENTS
Study Areas
The Lower White River Region (LWRR) study site is an 8,970 square km region in
eastern Arkansas (Figure 2b). The LWRR study site was selected such that it encompassed the
White River National Wildlife Refuge (NWR), Cache River NWR, Bald Knob NWR (Figure 9),
and a full range of land cover observed throughout the region. The final decision about the
extent of the LWRR study area was determined by the geographic availability of digital National
Wetland Inventory data (Table la). All LWRR habitat maps are overlaid on a grayscale Landsat
Enhanced Thematic Mapper (ETM+) satellite image (single near infra-red band, 30m spatial
resolution) displaying a minimum of an additional 1500 m region outside the study boundary.
The satellite image, county boundaries, reference cities, and federal wildlife refuge boundaries
may be used to identify specific locations on all LWRR maps, referencing Figure 2b.
The Mississippi Alluvial Valley Ecoregion (MAVE) study site is a 233,782 square km
region in the Lower Mississippi River Valley, which primarily intersects Louisiana, Mississippi,
Arkansas, and Missouri and small portions of Illinois and Kentucky (Figure 2). The MAVE
delineates a regional area that contains similar land uses, soils, land surface forms, and potential
natural vegetation types (Omernik, 1987). The MAVE contains the entire LWRR study area,
and overlaps with a portion of the WRW (Figure 2a and Figure 2c). All MAVE habitat maps
are overlaid on county and state boundaries, which may be used to identify specific locations
throughout the MAVE, referencing Figure 2c.
How to Navigate and Interpret Habitat Assessment Maps
Figure 11 depicts the two types of habitat assessment maps that appear in this chapter
along with a brief explanation of each map element. The habitat suitability maps depict the
different parameters for each of the selected taxa, in accordance with available literature for
mallard ducks (Table 2a), black bears (Table 2b), least terns (Table 2c), and wetland plants
(Table 2d). Habitat suitability maps are color-coded to best distinguish between habitat types
and patches, and are not intended to imply relative importance of habitat type(s) for any taxon.
Habitat vulnerability maps are consistently color-coded from greatest vulnerability (red tones) to
-------
least vulnerability (blue tones), based on standard deviations from the mean of a particular
vulnerability parameter, and are intended to imply the relative risk of loss or damage to each
individual patch of habitat for the taxon or guild indicated on the map (Figure 11).
Specific locations on all maps may be examined in detail with the aid of the zoom and
pan buttons in the icon bar if you are using the digital version of this atlas. For both the LWRR
and MAVE study areas a very small percentage (less than 4 percent by area) of the wetland
habitat assessed intersects with the study area boundary. These 'edge patches' have no effect on
the habitat suitability maps. The habitat patches that intersect the study area boundary exert a
minimal effect on the distribution of relative habitat vulnerability classes, which are based on a
mean vulnerability parameter (Figure 11).
Habitat Suitability
The ecological bases of the habitat suitability models in this atlas are summarized below.
The GIS-based habitat suitability models in this atlas use specific habitat requirements for
mallard ducks, black bears, least terns, and wetland plants, to the extent that digital data is
available to model their ecology in prior field research. The following ecological overview of
organisms is condensed to include the information relevant to the GIS models contained within
this atlas.
Mallard Duck Winter Habitat (summarized from Appendix A)
The Lower Mississippi Valley is the primary wintering habitat for mallard ducks in the
Mississippi Flyway (Bellrose, 1976), resulting in a residence of an estimated 1.6 million ducks
during the winter months (Bartonek et al., 1984). A wide variety of wetland hydrologic and
vegetational conditions are required to meet different habitat requirements of various mallard
sexes, ages, and behavioral segments of the mallard population. Generally, winter habitat
conditions in the Lower Mississippi region influence all aspects of the socio-biology of mallards,
which in turn affects the fecundity and survival of mallards. For example, mallard ducks in the
Lower Mississippi region typically move 1.6 km to 8 km from roost sites to foraging areas.
Flights of greater than 8 km are typically a response to changes in hydrology, temperature,
depleted food resources, or other disturbances (Jorde et al., 1983).
-------
Mallard ducks are omnivores that feed on available foods that include aquatic
invertebrates; wetland plant seeds, fruits, rootlets, and tubers; mast from trees; agricultural
grains; mollusks; insects; small fish; fish eggs; and amphibians (Heitmeyer, 1985; Allen, 1987;
Appendix A). Consequently, wetlands and other open water areas are extremely important for
mallard ducks because such areas are where most of the naturally occurring duck food occurs.
The timing of flooding, flood depth, and the duration of wetland flooding is a critical factor in
determining the diversity and availability of organisms upon which mallard ducks feed
(Fredrickson, 1979; Heitmeyer and Fredrickson, 1981; Nichols et al., 1983; Heitmeyer, 1985;
Allen, 1987), and much larger numbers of mallards have been observed in the Lower Mississippi
Valley when these wetter conditions exist (Nichols et al., 1983), particularly in the forested
wetlands (Heitmeyer, 1985). Mallards typically feed from the water's surface to a maximum of
50 cm (Heitmeyer and Fredrickson, 1981; Krapu, 1981; Batema et al., 1985; Allen, 1987). Thus,
the depth of flooding, the duration of flooding, and the type of vegetation, as it relates to food
resources, is cited as the primary determiner of mallard duck habitat suitability (Allen, 1987).
Accordingly, these factors were used to establish suitable mallard habitat and to distinguish
between mallard habitat patches of differing suitability characteristics (Table 2a). Although
some of the agricultural land in the LWRR is managed to provide winter mallard forage,
agricultural grains are not a complete substitute for natural foods (Frederickson and Taylor,
1982; Baldassarre et al., 1983; Jorde et al., 1984). Therefore, agricultural land was not
considered to be suitable mallard habitat relative to non-agricultural wetland and was not used in
the habitat vulnerability assessments in this atlas. A hierarchical schematic of the mallard duck
habitat suitability model construction process is shown in Figure 12, depicting the links between
the original digital data, the habitat suitability model, and the habitat vulnerability models. Black
bear GIS models and wetland plant GIS models follow processes that are similar to the mallard
duck GIS models, but use different GIS data layers.
Because all of the naturally occurring and diverse habitat requirements of mallard ducks
occur in wetlands, mallards require access to a variety of wetland types, including emergent
wetlands (Figure 13), scrub/shrub wetlands (Figure 14), forested wetlands (Figure 15),
unconsolidated bottom wetlands (Figure 16), aquatic bed wetlands (Figure 17), and the open
water areas of lakes, impoundments, or rivers (Figure 18). The wetland types described in this
atlas are based on the wetland classes and hydrologic modifiers used in the National Wetland
-------
Inventory (NWI; after Cowardin et al., 1979), as they apply to mallard duck habitat requirements
(Figure 19). Specifically, mallard duck winter habitat suitability models in the LWRR and
MAVE are based on optimal foraging-related habitat requirements of: (a) available wetland
habitat; (b) the hydroperiod within a wetland (i.e., either temporarily-flooded/seasonally-flooded,
permanently flooded, or dry); (c) the presence of woody vegetation; and (d) the presence of
desirable mast producing oaks, specifically excluding overcup oak (Quercus lyratd) which
produce acorns up to 2.5 cm long and are thus less suitable oak mast for forage (Allen, 1987;
Table 2a). We chose mallard ducks as one of the modeled taxa for this atlas because they are
abundant and ubiquitous, are dependent upon wetlands for most of the year, have well
documented habitat requirements, require habitat that can be readily mapped using GIS data, and
are a species that has recovered from previously lower numbers in the study area and throughout
the Mississippi Flyway.
Black Bear Wetland Habitat (summarized from Appendix B)
Black bears are found throughout North America and, in the LWRR and MAVE, have a
mean female home range from 9 km2 to 12 km2. Male home ranges have been reported as large
as 116 km2 to 148 km2 (Pelton, 1982; Klepinger and Norton, 1983; Smith, 1985), while others
have reported male home ranges from 13 km2 to 24 km2 (Rogers, 1992). The home ranges of
male and female black bears are primarily dependent on the availability of resources (Jonkel and
Cowan, 1971; Amstrup and Beecham, 1976; Garshelis and Pelton, 1980; LeCount, 1980;
Reynolds and Beecham, 1980; McArthur, 1981; Elowe, 1984; Rogers, 1987), and are also
influenced by population density, age of the individual, and seasonal conditions (Pelton, 1982).
In 1997, one hundred eighty-seven black bears were legally killed in Arkansas using
muzzleloader, modern gun, archery, and crossbow hunting techniques, solely in the mountainous
regions. As a result of increasing numbers of black bears in the vicinity of the LWRR, hunting
of black bears within selected areas to the west of the Mississippi River, using modern guns,
became legal in 1999 excluding the White River National Wildlife Refuge (Arkansas Game and
Fish Commission, 1998).
Black bears are omnivores that typically feed on easily digestible vegetative foods that
are high in nutrients and low in cellulose (Rogers, 1976; Herrero, 1978; Herrero, 1979; Rogers,
1987). Thus, the typical black bear diet consists of fruits, nuts, acorns, insects, and early-
-------
sprouting green vegetation (Mealey, 1975; Herrero, 1979; Rogers, 1987). When naturally
growing food items are scarce, black bears may alternatively feed on agricultural crops, such as
orchard fruits or corn or at human-constructed food sources, such as centralized refuse disposal
sites or at the residences of humans (Harger, 1967; Bray, 1974; Rogers, 1976; Rogers et al.,
1976; Hugie, 1979; Landers et al., 1979; Beeman and Pelton, 1980; Rogers, 1987). Scarcity of
naturally growing foods for black bears has been positively correlated with occurrences of bear
cannibalism (Tietje et al., 1986) and the occurrences of bear-human interactions. Black bears in
the LWRR and MAVE may remain active throughout the winter to feed on corn and other foods
if other naturally growing foods are scarce (Carpenter, 1973; Matula, 1974; Lindzey et al., 1976;
Rogers, 1976; Hamilton, 1978; Hamilton and Marchington, 1980; Elowe, 1984). Black bear
predation on vertebrates is relatively rare but such captures in the LWRR and MAVE may
include newborn deer, nestling birds, fish, or other animals whose escape is hampered (Rowan,
1928; Barmore and Stradley, 1971; Frame, 1974; Cardoza, 1976; Ozoga and Verme, 1984).
Wetlands and other open water areas are extremely important for black bears to survive
because, aside from providing much of the food resources that they require, such areas are
frequently the sole resource for drinking water and for providing water in which they can cool
themselves. The home ranges of black bears are also closely tied to forested areas (Herrero,
1979; Hugie, 1979; Pelton, 1982), and are limited by the fact that much of the remaining forested
areas in the LWRR and MAVE are fragmented and therefore relatively inaccessible (Cowan,
1972, Maehr and Brady, 1984; Twedt et al., 1999). Black bears in relatively fragmented
landscapes, like the MAVE, are frequently observed in forest openings and clearings, which may
provide a relatively higher degree of edge vegetation diversity than core forest areas (Herrero,
1979; Hugie, 1979). Wetlands, particularly in riparian areas, are used by black bears for
seasonal foraging; denning; cover for escape; and as travel corridors. Thus, as a result of the
diverse resource and travel corridor requirements, black bears require access to a variety of
wetland types and resources, including emergent wetlands (Figure 13), scrub/shrub wetlands
(Figure 14), forested wetlands (Figure 15), unconsolidated bottom wetlands (Figure 16), aquatic
bed wetlands (Figure 17), and the open water areas of lakes, impoundments, or rivers (Figure
18). Specifically, the black bear habitat suitability models in the LWRR and MAVE are based
on optimal habitat requirements of: (a) available wetland habitat; (b) the presence of woody
vegetation; (c) the number of plant species within a patch, and (d) evidence of forest disturbance
-------
since the 1950s (Table 2b). We chose black bears as one of the modeled taxa for this atlas
because they are a recovering species in the region, are dependent upon wetlands for most of the
year, have well documented habitat requirements, and require habitat that can be readily mapped
using GIS data.
Least Tern Breeding Habitat (summarized from Appendix C)
Least terns breed along marine and freshwater coastal areas of North, Central, and South
America and the Caribbean Islands (American Ornithologists' Union, 1983). There are three
subspecies of least terns in the United States, including the interior least tern (Sterna antillarum
athalassos\ which is modeled in this atlas. The interior least tern breeds along the major
tributaries of the Mississippi River (Ducey, 1981; Cobb, 1992; U.S. ACOE, 1999) and within the
Rio Grande River watersheds (Downing, 1980). The U.S. Fish and Wildlife Service lists the
least tern (Sterna antillarum) as a federally endangered species (Endangered Species Act, 1973).
The other two subspecies of least terns, not modeled in this atlas, are the eastern least tern (S. a.
antillarum) that breeds along the Atlantic and Gulf of Mexico coasts (American Ornithologists'
Union, 1983), and the California least tern (S. a. brownf) that breeds from San Francisco Bay to
southern Baja California, Mexico (California Least Tern Recovery Team, 1980).
Least tern breeding habitat is generally characterized as open sand, soil, or dried mud in
the proximity of water (Hardy, 1957; Craig, 1971; Massey, 1971; Massey and Atwood, 1982;
Appendix C). Foraging areas of interior least terns include rivers, lakes, ponds, sloughs, and
borrow pits (Ganier, 1930). Foraging areas of least terns are usually close to breeding areas,
with typical ranges of breeding individuals reported from 100 m to 1.6 km, and as far away as
6.4 km (Jernigan et al., 1978; Hays, 1980; Massey and Atwood, 1981; Atwood and Minsky,
1983; Faanes, 1983; Carreker, 1985). Least terns tend to inhabit ephemeral sandy shorelines
year after year (Burger, 1984) even if reproduction has declined in prior years at these locations
(Massey and Atwood, 1979). Site intolerance among least terns is related to vegetation
encroachment (Burger, 1984); beach erosion (Downing, 1973); human-related disturbances that
result from replacement of natural land cover with built structures (Chambers, 1908; Massey and
Atwood, 1980; Ducey, 1981; Grochfeld, 1983); replacement of natural land cover with
agricultural land (Schulenburg and Ptacek, 1984); and river channel deepening (Downing, 1980).
Least terns are predominantly fish-eating birds, hovering and diving from 3 m to 10 m above the
-------
surface of the water (Hardy, 1957; Tomkins, 1959; Moseley, 1976). Least terns may also skim
the surface of the water for prey (Bent, 1921; Oberholser, 1974) and occasionally feed on insects
over land (Bent, 1921; McDaniel and McDaniel, 1963; Moseley, 1976; Schulenberg et al., 1980).
In general, adult least terns consume fish in the 2.5 cm to 9 cm length range (Massey, 1974;
Moseley, 1976; Massey and Atwood, 1980; Schulenberg et al., 1980). Accordingly, least tern
habitat suitability models in the LWRR and MAVE are based on the optimal habitat
requirements of available bare shoreline of lakes, impoundments, and rivers (Figure 20) during
the nesting season from June through August (Table 2c). The relative rarity of bare shoreline
compared to other land cover types in the LWRR and MAVE eliminated the necessity for further
division of habitat characteristics in this atlas, such as the proximity of breeding areas to foraging
habitat. We chose least terns as one of the modeled taxa for this atlas because they are an
endangered species, dependent upon wetlands and aquatic ecosystems for most of the year, have
well documented habitat requirements, and require habitat that can be readily mapped using GIS
data.
Wetland Plant Habitat
Wetland plants are adapted to surviving in soils that are saturated with water (and
consequently have a low oxygen content). Plants that flourish in wetlands are thus generally
referred to as hydrophytes, but may also include plants that can survive with only brief periods of
soil saturation, and the low-oxygen soil conditions that accompany soil saturation (Reed, 1988;
Appendix D). Wetland plant species in the LWRR and the MAVE are numerous and provide
cover and forage for other organisms in emergent wetlands (Figure 13), scrub/shrub wetlands
(Figure 14), forested wetlands (Figure 15), aquatic bed wetlands (Figure 17), and in the littoral
zones of some lakes, impoundments, and rivers (Figure 18). Accordingly, wetland plant habitat
suitability models in the LWRR and MAVE are based on the optimal wetland plant habitat
requirements of: (a) the presence of wetland conditions (particularly, the presence of wetland
soil types; Mitsch and Gosselink, 1993) and (b) the hydroperiod within each wetland type (Table
2d). We chose the collective group (i.e., a guild) of regional wetland plant species for this study
because they are strictly limited to the geographic extent of wetlands, are strictly dependent upon
wetland hydrology, have well documented habitat requirements, have well understood
-------
physiological responses to hydrologic and other physical disturbances, and require habitat that
can be readily mapped using GIS data.
Habitat Suitability Maps
Wetland and upland habitat suitability is mapped for mallard ducks (LWRR, Figure 2la;
MAVE, Figure 24a), black bears (LWRR, Figure 21b; MAVE, Figure 24b), and least terns
(LWRR, Figure 21c and Figure 2Id; MAVE, Figure 24c and Figure 24d). Maps that solely
include wetland habitat are displayed for mallard ducks (LWRR, Figure 21e; MAVE, Figure
24e), black bears (LWRR, Figure 2If; MAVE, Figure 24f), and wetland plants (LWRR, Figure
2 Ig; MAVE, Figure 24g).
Wetland habitat patches less than or equal to 2 ha were mapped separately for mallard
ducks (LWRR, Figure 21h; MAVE, Figure 24h), black bears (LWRR, Figure 21i; MAVE,
Figure 24i), and wetland plants (LWRR, Figure 21j; MAVE, Figure 24j) because these smaller
patches of wetland habitat account for 20,093 cumulative ha of wetlands in the LWRR and
196,360 cumulative ha of wetlands in the MAVE. These smaller wetlands are frequently
overlooked, individually and collectively, in habitat assessments (Klett and Kirsch, 1976;
Robinson, 1995; Naugle et al., 2002; Stevens et al., 2002). Thus, we included habitat suitability
for patches less than or equal to 2 ha, but did not measure habitat patch vulnerability further in
these smaller patches because of the accuracy limitations of the GIS data and the minimum
mapping unit for each model (Table la and Table Ib). Suitable wetland habitat patches greater
than 2 ha are mapped for mallard ducks (LWRR, Figure 21k; MAVE, Figure 24k), black bears
(LWRR, Figure 211; MAVE, Figure 241), and wetland plants (LWRR, Figure 21m; MAVE,
Figure 24m). The habitat suitability maps provide the bases for the habitat vulnerability models
in the remainder of this atlas.
Habitat Vulnerability
Because the GIS models in this atlas are based on 30 m resolution data (e.g., AR-GAP in
the LWRR; NLCD in the MAVE), caution should be exercised when using these results for fine
scale interpretation (see Habitat Suitability section regarding habitat patches of 2 ha or less). The
edge of each habitat patch was defined by a change in the wetland habitat suitability class for a
particular taxa or guild (i.e., wetland plants). Thus, habitat vulnerability was determined for all
-------
suitable wetland habitat patches in the LWRR and MAVE for mallard duck, black bear, and
wetland plants if a habitat patch was greater than 2 ha. All least tern habitat is extremely rare,
and thus extremely vulnerable to loss or damage (LWRR, Figure 21c and Figure 2Id; MAVE,
Figure 24c and Figure 24d). Because of the relative rarity and minimal total area of least tern
habitat, vulnerability measures were not calculated for their habitat patches.
Habitat vulnerability measures (Table 3a) within habitat patches for each taxa are based
on patch size, patch shape, human-induced disturbances, and combinations of these metrics. All
of the habitat vulnerability metrics within a habitat patch may affect the likelihood that a
particular habitat patch will rebound after patch disturbance(s). That is, habitat vulnerability
metrics are based on predicted habitat degradation as a result of patch destruction (i.e., total
loss), patch fragmentation (i.e., partial loss), or patch degradation (i.e., stress) [after Odum,
1985).
Patch size metrics in this atlas (Table 3a) are based on previously observed ecosystem
trends regarding the effects of patch size on habitat quality for specific taxa, in many different
regions (e.g., MacArthur and Wilson, 1967; Simberloff and Wilson, 1970; Diamond, 1974;
Forman et al., 1976; Pickett and Thompson, 1978; Soule et al., 1979; Hermy and Stieperaere,
1981; van der Valk, 1981; Simberloff and Abele, 1982; McDonnell and Stiles, 1983; Harris,
1984; McDonnell, 1984; Moller and Rordam, 1985; Brown and Dinsmore, 1986; Dzwonko, and
Loster, 1988; Gutzwiller and Anderson, 1992; Opdam et al., 1993; Hamazaki, 1996; Kellman,
1996; Bastin and Thomas, 1999; Mclntyre and Wiens, 1999a; Mclntyre and Wiens, 1999b;
Twedt and Loesch, 1999; Jones et al., 2000b; Lopez et al., 2002; Lopez and Fennessy, 2002).
Accordingly, the 'patch size' habitat vulnerability models map the habitat patch area, habitat
patch perimeter length, and habitat patch interior-to-edge ratio in the LWRR (Figure 22) and the
MAVE (Figure 25). Smaller habitat patches (as measured by area, perimeter, or interior-to-edge
ratio) are relatively less likely to rebound from disturbances (i.e., are more likely to be
fragmented or destroyed after changes in hydrology, destruction of vegetation, or the
establishment of opportunistic flora and fauna) than larger habitat patches.
The ecological vulnerability metrics in this atlas regarding patch shape (Table 3a) are
based on trends previously observed for specific taxa, in many different regions (e.g., MacArthur
and Wilson, 1967; Gilpin, 1981; McDonnell and Stiles, 1983; Harris, 1984; McDonnell, 1984;
Gutzwiller and Anderson, 1992; Hamazaki, 1996; Kellman, 1996; Bastin and Thomas, 1999;
-------
Jones et al., 2000b; Lopez et al., 2002; Lopez and Fennessy, 2002), and include patch 'sinuosity'
(after Bosch, 1978; Davis, 1986) in the LWRR (Figure 22) and the MAVE (Figure 25). Patches
with a smaller sinuosity index are less winding or convoluted in shape, thus are relatively less
likely to rebound from disturbances than patches with a greater sinuosity index. That is, smaller
index values indicate that a habitat patch has a greater likelihood of fragmentation or loss as a
result of environmental change, such as changes in hydrology, destruction of vegetation, or the
establishment of opportunistic flora and fauna. Another metric of patch shape is patch
'circularity' (after Stoddart, 1965; Unwin, 1981) in the LWRR (Figure 22) and the MAVE
(Figure 25). Habitat patches with a smaller circularity index are more circular in shape, thus are
relatively less likely to rebound from disturbances than patches with a larger circularity index.
That is, smaller index values indicate that a habitat patch has a greater likelihood of
fragmentation or loss as a result of environmental change, such as changes in hydrology,
destruction of vegetation, or the establishment of opportunistic flora and fauna. The unified
patch index (Table 3a) multiplicatively combines the patch interior-to-edge metric, the patch
sinuosity index, and the patch circularity index into a single metric that depicts patch area, patch
perimeter, and patch shape simultaneously as a measure of habitat patch vulnerability in the
LWRR (Figure 22) and the MAVE (Figure 25).
Human-induced disturbance factors within habitat patches in the LWRR and MAVE
(Table 3a) are based on previously observed positive correlations between ecosystem
degradation and amount of land cover conversion during road construction, road maintenance,
and other human-activities (e.g., Connell and Slatyer, 1977; van der Valk, 1981; Ehrenfeld,
1983; Johnston, 1989; Scott et al., 1993; Johnston, 1994; Poiani and Dixon, 1995; Strittholt and
Boerner, 1995; Jenning, 1995; Wilcox, 1995; Ogutu, 1996; Stiling, 1996; Heggem et al., 2000;
Lopez et al., 2002; Lopez and Fennessy, 2002). Thus, the directly measurable human-induced
disturbance metrics within habitat patches are road length and road density in the LWRR (Figure
23) and the MAVE (Figure 26). Patches with greater total road length or road density (or their
combined index value; Table 3a) are relatively less likely to rebound from disturbances than
patches with a lesser road presence. That is, greater road length, road density, or road index
values indicate that a patch has a greater likelihood of fragmentation or loss as a result of
environmental change, such as changes in hydrology, destruction of vegetation, or the
establishment of opportunistic flora and fauna. The increased presence of roads within a patch
-------
may also bring about a decrease in patch size or shape metrics. The road index additively
combines road length and road density metrics to depict the combined potential habitat
degradation effects of road construction and maintenance on habitat patches in the LWRR
(Figure 23) and the MAVE (Figure 26).
The indirect metrics of human-induced disturbance within habitat patches are human
population density in 1990, estimated human population density in 2011 (i.e., future human
population density), and estimated human population density change from 1990 to 2011 for the
LWRR (Figure 23) and the MAVE (Figure 26). The human population density change metric
was normalized to a positive number by adding 50 to the calculation for the LWRR, and by
adding 8000 to the calculation for the MAVE models (Table 3a). Habitat patches that exist
within in census block groups (Table la and Table Ib) with a greater population density now, or
in the future are relatively less likely to rebound from disturbances than patches that exist in
areas of lesser population density, now or in the future. That is, greater human population
density values indicate that a habitat patch has a greater likelihood of fragmentation or loss as a
result of environmental change, such as changes in hydrology, destruction of vegetation, or the
establishment of opportunistic flora and fauna. The increased presence of humans near habitat
patches may also bring about a decrease in patch size metrics, patch shape metrics, an increase in
road length, or an increase in road density. The unified human index additively combines road
length, road density, and human population density change (from 1990 to 2011) to depict the
combined effects of direct and indirect human-induced disturbance in the LWRR (Figure 23) and
the MAVE (Figure 26). The unified vulnerability index (Table 3a) additively combines the
unified patch index, direct human-induced disturbance metrics, and indirect human-induced
disturbance metrics, so that the combined effects of patch size, patch shape, road metrics, and
human population density metrics can be depicted simultaneously in the LWRR (Figure 23) and
the MAVE (Figure 26). Applications of the habitat vulnerability metrics to wildlife refuges and
for landscape simulations are described in Chapter 4.
-------
CHAPTER 3. WATER QUALITY VULNERABILITY ASSESSMENTS
Study Area
The White River Watershed (WRW) study site is a 101,533 square km region in
Arkansas and southern Missouri (Figure 2a and Figure 2c). The study site was selected such that
it encompassed the twenty-five 8-digit HUC subwatersheds (Figure 27) comprising the entire
catchment area of the White River, and contained a full range of the land cover types in the
region (Figure 3). Thirty-eight National Water Quality Assessment (NAWQA) surface water
sampling locations are located within the WRW (Figure 27). Field-based water quality data at
the thirty-eight NAWQA locations were used to validate some of the metrics described in this
chapter, and were sampled from 1992 through 2000 in the Spring and Summer months (U.S. GS,
2001). All land cover metric maps in this atlas are overlaid on 8-digit HUC subwatershed
reference boundaries (Figure 27) so that they may be referenced in relationship to the locations
of counties in the WRW (Figure 2c). Additional information about a specific watershed may be
accessed at the following URL: http://www.epa.gov/surf/.
How to Navigate and Interpret Water Quality Vulnerability Maps
Figure 11 illustrates the types of watershed-based land cover maps that appear in this
chapter. Using the land cover types provided in the National Land Cover Dataset (NLCD; Table
4), several land cover metrics were developed to assess the vulnerability of surface water to
pollution from various land cover types and configurations (Table 3b), including the proximity of
land cover to surface water. Measurements of land cover in proximity to surface water were
calculated by intersecting a 'buffer zone' around each surface water body (Figure 10) with each
land cover metric (Table 3b). These measurements were then used to create a watershed map
that depicts the percent of a selected land cover metric at the bank, and within cumulative 30 m
buffer zone intervals (Figure 11) from all water bodies. For each map, the data range in the
legend shows the metric values within each of the 8-digit HUC subwatersheds, contained in each
of five quintiles. A quintile contains one-fifth of the distribution of each watershed metric.
Quintiles are formed after ranking watersheds for each metric. The map of the WRW is color-
coded to show relative conditions among watersheds, ranging from red (i.e., greatest
-------
vulnerability) to green (i.e., least vulnerability), potentially indicating a relative range in terms of
surface water quality impairment from nutrient or sediment loading.
Water Quality Vulnerability
Land-cover gradients within the White River Watershed were used to develop landscape
indicators of surface water quality at a broad scale. The theoretical relationships between
watershed land cover and the surface water chemistry within the WRW are based on previously
published field results. For example, there is a strong negative correlation between the presence
of 'natural vegetation' (i.e., trees, grassland, and wetland vegetation) and the amount of
nutrients (e.g., nitrogen and phosphorus) and sediment in downstream surface water (Gregory et
al., 1991; Valett et al., 2002). Peterjohn and Correll (1984) demonstrated that the negative
correlation between the presence of natural vegetation in riparian zones and nutrient loading
from agricultural land to streams is primarily a result of nutrient uptake in the root zone, and
assimilation of these nutrients into vegetative plant parts. Other research strongly suggests that
the high degree of clay and organic components in wetland soils, anoxic soil conditions, and low
through-flow in wetlands (in combination with natural vegetation) are responsible for the strong
negative correlations observed between the presence of 'natural areas' and the concentrations of
nutrients, sediment, or trace metals in connected surface water bodies (e.g., Hey et al., 1989;
Johnston, 1989; Gregory et al., 1991; Poiani et al., 1996; Fennessy and Cronk, 1997; Crumpton
and Baker; 19981 Giese et al., 2000). Historically, the loss of forests and other natural vegetation
in the MAVE is a result of the expansion of agricultural land development (see Chapter 1) and
there is a strong positive correlation between the presence of agricultural land cover and the
amount of nutrients and sediment in downstream surface water bodies (e.g., Johnston et al.,
1965; Triplett et al., 1969; Romkens et al., 1973; Hergert et al., 1981; Robinson and Sharpley,
1995). Therefore, we selected road, agriculture, and natural vegetation metrics (NLCD codes 31,
41, 42, 43, 51, 71, 91, 92 in Table 4) as potential landscape indicators of surface water quality,
and mapped these metrics among HUC subwatersheds in the WRW (Figure 27).
Roads and Agriculture
There are 132,042 kilometers of streams and 130,165 kilometers of roads in the WRW.
Information about the location of surface water in the WRW, including the location of streams,
-------
rivers, lakes, impoundments, and ditches (Figure 28) was used in combination with information
about the location of roads (Figure 29) to determine the number of streams within thirty meters
of roads within the WRW (Figure 30). The placement of roads in close proximity to streams is a
surface water quality impairment risk because runoff from paved surfaces flows relatively
quickly to low areas (e.g., streams or lakes) and carries with it any substances that spill onto
roads. These substances can include hydrocarbons, metals, or other chemicals that are toxic to
plants or animals (Warren, 1981; Carter, 1982; Merian, 1990; Ehrenfeld and Schneider, 1991).
Information about the location of current land cover in the WRW from the NLCD (Figure
3; Table 4) was used to determine the percent of all agriculture (Figure 31; NLCD codes 81, 82,
83, and 85 in Table 4), percent crop agriculture (Figure 32; NLCD codes 82, 83, and 85 in Table
4), and percent agricultural pasture (Figure 33; NLCD code 81 in Table 4) within each HUC
subwatershed (Figure 27). Greater amounts of agricultural activities within a HUC subwatershed
increase the risks of surface water quality impairment because of the greater potential for soil
loss from tilled land, runoff of agricultural fertilizers and pesticides, or the runoff of livestock
byproducts. Modifying tilling practices may mitigate soil loss and runoff from crop-agricultural
land, because less soil is disturbed during plowing.
Information about the slope of the terrain within the WRW (Figure 34) was used in
combination with the land cover information from the NLCD to determine the percent of area
within each HUC subwatershed that contains agriculture on slopes greater than three percent
(Figure 35). Agricultural activities on steep slopes are a surface water quality impairment risk
because of the increased likelihood of soil erosion and loss to downhill areas and streams. Soil
loss and runoff on steep slopes may be mitigated by soil conservation practices such as terracing
pastures and fields parallel to the elevational contours of the land.
We selected 'percent total agriculture' for validity testing, using the 1990s NAWQA field
data, because it is an easily measurable variable that is relatively unchanging throughout the
WRW, thus repeatable with other available data sets. Significant positive correlations between
percent total agriculture and surface water concentrations of dissolved organic carbon (DOC),
amino and organic nitrogen (OrgN), total phosphorus (P), and suspended sediment (Sed) [Table
5] suggest that 'percent total agriculture' is an appropriate landscape indicator of actual increases
in nutrient and sediment concentrations in surface water of the WRW. The same significant
positive correlation trend exists for 'percent total agriculture' within the riparian zone, up
-------
through 300 m from shorelines (Figure 31). The significant positive correlations between
surface water chemistry measurements and streamside percent total agriculture (P < 0.0001)
suggests that a GIS data set containing solely the agricultural land cover data along streams is
sufficient to initially detect risk of surface water impairment from nutrient and sediment loading
as a result of agriculture in the WRW. These initial water quality vulnerability results, in
combination with a map analysis of percent total agriculture, among HUCs (Figure 36), suggest
that HUC 8020204 and HUC 8020201 are extremely vulnerable to water quality impairment as a
result of agricultural activities in proximity to surface water, particularly within 120 m of
shorelines (Figure 27).
Natural Vegetation
Information about the location of current land cover in the WRW from the NLCD was
used to create six landscape forest metrics (Figure 37), measuring 'largest 'forest patch
proportion of HUC', 'mean forest patch area', 'largest forest patch proportion of HUC', 'forest
patch density', 'forest patch number', and 'percent forest' (areal measurement) within a HUC.
The 'percent forest' metric (NLCD codes 41, 42, and 43 in Table 4) was compared among HUCs
and within cumulative 30 m riparian zones (Figure 38). Replacement of forests with other land
cover types is a surface water quality impairment risk because the root systems of trees and
forest understory vegetation tend to absorb substances that are plant nutrients (e.g., nitrogen) and
accumulate these substances in vegetation. Riparian forests also decrease the rate of overland
flow, and intercept a substantial amount of nutrients (e.g., phosphorus) and sediment,
incorporating these substances into wetland soils for long periods of time. Thus, 'percent
wetland' (including forested and emergent wetland vegetation; NLCD codes 91 and 92 in Table
4) was compared among HUCs and within cumulative 30 m riparian zones (Figure 39). All
natural land cover types in the NLCD (i.e., NLCD codes 31, 41, 42, 43, 51, 71, 91, and 92 in
Table 4) were combined to measure the percent of natural land cover in the landscape, and were
compared among HUCs and within cumulative 30 m riparian zones (Figure 40).
Although many of the theoretical 'natural vegetation loss' risks of surface water quality
impairment are important, we selected 'mean forest patch area', 'largest forest patch proportion
of HUC', and 'percent forest' for validity testing, using the 1990s NAWQA field data, because
these metrics are easily measurable using existing GIS data sets. Significant positive correlations
-------
between the selected landscape forest metrics and the surface water concentrations of DOC,
OrgN, P, and Sed (Table 6) suggest that 'mean forest patch area', 'largest forest patch proportion
of HUC', and 'percent forest' are appropriate landscape indicators of actual increases in nutrient
and sediment concentrations in surface water of the WRW. The same significant positive
correlation trend exists for 'percent forest' within riparian zone distances, up through 300 m
from shorelines (Figure 38). The significant positive correlations between surface water
chemistry measurements, 'mean forest patch area', 'largest forest patch proportion of HUC', and
either HUC or streamside percent forest (P < 0.0001) suggests that a GIS data set containing
solely forest cover data is sufficient to detect risk of surface water impairment from nutrient or
sediment loading in the WRW. These initial water quality vulnerability results, in combination
with a map analysis of percent forest among HUCs, suggest that six HUCs are extremely
vulnerable to water quality impairment as a result of forest loss in proximity to streams, with
HUC 11010014 as the subwatershed with the greatest loss of forest on land that is directly
adjacent to streams (Figure 41).
-------
CHAPTER 4. ASSESSMENT OF HABITAT
VULNERABILITY IN WILDLIFE REFUGES
National Wildlife Refuges in the Lower White River Region
Mallard duck winter habitat vulnerability models were used to assess the 895 square
kilometers of federal refuge lands within the 8,921 square kilometer Lower White River Region
(LWRR). LWRR mallard duck winter habitat (Table 7) is contained within thirteen Arkansas
counties (Figure 2b), of which a portion intersects the seventy-two separate parcels of land that
comprise the White River National Wildlife Refuge (NWR), Cache River NWR, and Bald Knob
NWR (Figure 9 and Figure 42). The U.S. Fish and Wildlife Service (DeWitt, Arkansas, 2001)
supplied the boundary of each National Wildlife Refuge Zone (RZ, hereafter) as Arc View shape
files (Environmental Systems Research Institute, v.3.2, Redlands, California).
Results
The 89,529 ha of federal refuge land in the LWRR ranges in size from 7.6 ha (RZ 12) to
64,552 ha (RZ 72), with a mean RZ area of 2,558 ha (S.D. = 10,838 ha; Table 8). Refuge
perimeter length ranges from 1,182 m (RZ 12) to 335 km (RZ 72), with a mean RZ perimeter
length of 25 km (S.D. = 55 km). RZ area is strongly negatively correlated with mean percent
contribution of habitat patches within the RZ (p = - 0.723, P = <0.0001) and weakly positively
correlated with mean area of habitat patches within the RZ (p = 0.127, P = <0.0001). To
examine the influence of the four largest RZs (contributing 83% of the area to all RZs in the
LWRR) we excluded them (i.e., RZs 72, 67, 62, and 8) from the analyses (Table 9) but
correlations between habitat area and refuge area did not substantially change (Table 9). The
rank order of RZs with regard to RZ area, RZ perimeter, area of habitat patches within RZ, and
mean percent contribution of habitat patches within a RZ suggests that larger RZs tend to capture
larger wetland habitat patches, but this relationship is nonlinear, and the relationship is most
clearly demonstrated among either relatively larger RZs or relatively smaller RZs (Table 10).
Eight specific RZs demonstrate how a larger refuge does not necessarily result in larger mallard
habitat within the refuge. RZ 29 is relatively large with very small intersecting habitat patches;
and RZs 48, 44, 40, 32, 18, 13, and 11 are relatively small with large intersecting habitat patches.
-------
Table 9 summarizes the general trend that RZ area is weakly negatively correlated with
(a) original area of habitat patches inside and outside of the RZ (p = - 0.141, P = <0.0001); (b)
habitat patch perimeter within a RZ (p = - 0.095, P = <0.0001); (c) habitat patch interior-to-edge
ratio within a RZ (p = - 0.188, P = <0.0001); (d) population density in the year 1990 (p = -
0.214, P = O.0001); (e) estimated population density in the year 2011 (p = - 0.233, P =
<0.0001); and (f) estimated population density change from 1990 to 2011 in a RZ (p = - 0.039, P
= <0.031). To hold aside the influence of the four largest RZs we excluded them from the metric
comparisons, resulting in trends similar to the full group of seventy-two RZs (Table 9).
Rankings of mallard duck wetland habitat patch metrics and indexes (Table 11) indicate that (a)
RZs 72, 67, 62, and 8 are relatively more vulnerable to disturbance as a result of a smaller habitat
patch interior-to-edge ratio; (b) RZs 72, 67, and 8 are relatively more vulnerable to disturbance
as a result of a smaller habitat patch area; (c) RZs 72 and 8 are relatively more vulnerable to
disturbance as a result of a smaller habitat patch perimeter, sinuosity index, and circularity index;
and (d) RZ 8 is relatively more vulnerable to disturbance as a result of a smaller habitat patch
unified patch index (Table 11).
Table 12 describes the seventy-two RZs, ranked by 'human-induced disturbance' metrics
and indexes, with the median indicated for each value range. The ranks of RZs indicate that (a)
RZs 72, 67, 62, and 8 are relatively more vulnerable to disturbance as a result of a greater habitat
patch road density, road length, road index, and unified human index; (b) RZs 67, 62, and 8 are
relatively more vulnerable to disturbance as a result of a greater habitat patch population density
in 1990 and expected population density in 2011; (c) RZ 8 is relatively more vulnerable to
disturbance as a result of a greater habitat patch population density change expected in the year
2011; and (d) RZs 72, 67, 62, and 8 are relatively less vulnerable to disturbance as indicated by
the relatively larger unified vulnerability index (Table 12).
For wetland habitat patches less than or equal to 2 ha there is a greater percent
contribution of the original habitat patch area in smaller RZs (by area, p = -0.897, P < 0.0001; by
perimeter, p = -0.868, P < 0.0001) than in larger RZs. The ranks of RZs with regard to wetland
habitat patches less than or equal to 2 ha within them indicate that all RZs have patches that are
less than or equal to 2 ha, with the exception of RZs 31, 28, 22 and 2 (Table 13; Figure 42;
Figure 9). For wetland habitat patches less than or equal to 2 ha the mean original habitat patch
area is weakly positively correlated with the area of RZ (p = 0.285, P = 0.0197) and the
-------
perimeter of the RZ (p = 0.310, P = 0.0112). For wetland habitat patches less than or equal to 2
ha, the strongest positive correlation exists between total habitat patch count and the area of RZ
(p = 0.909, P < 0.0001) and the perimeter of the RZ (p = 0.905, P < 0.0001). Because wetlands
that are 2 ha or smaller are too small to reliably monitor with currently available satellite remote-
sensing-derived land cover products, routine broad-scale monitoring of such small wetland areas
may be difficult.
Discussion
RZ gradient analyses in the LWRR suggest that larger habitat patches are relatively more
likely to rebound (i.e., are less likely to be fragmented or destroyed) than smaller habitat patches
after disturbances, such as changes in hydrology, destruction of vegetation, or the establishment
of opportunistic flora and fauna. However, large size alone may not be sufficient to ensure that a
patch is capable of existing and flourishing in a disturbed setting; other contributing disturbance
factors such as patch perimeter length, interior-to-edge ratio, sinuosity, and circularity may also
be relevant (see Chapter 2). Additionally, human-induced disturbances, such as runoff,
agriculture, and land conversion may function as drivers of habitat degradation, fragmentation,
and loss. Thus, those areas with increased human-induced disturbance (as evidenced by present
and future population density, population density change, and road development) in the LWRR
may bring about future net wetland degradation or loss in the LWRR.
The positive correlation between wetland habitat area or the percent contribution of
habitat area, with area of RZ indicates that larger refuges contain more wetland habitat than
smaller refuges. However, the percent contribution of wetland habitat to a refuge is inversely
correlated with the area of the refuge. The weakness of the correlation between RZ area and
mean area of habitat patches within a RZ, and outside a RZ, (Table 9) indicates that the presence
of a relatively large refuge is not necessarily a predictor of large habitat area within the refuge.
The weak negative correlation between RZ area, and the original area of the habitat patches
inside and outside the RZ; the habitat patch perimeter within a RZ; and the habitat patch interior-
to-edge ratio within a RZ (Table 11) also indicates that relatively large available and suitable
mallard duck winter habitat patches in the LWRR are not encompassed within current federal
refuge boundaries.
-------
The weak negative correlations between RZ area and human population density metrics
in the LWRR indicate that larger RZs tend to exist in areas of lower human population densities,
and smaller RZs tend to exist in areas of higher human population densities. Results of
estimated future human population density change in the RZs (from 1990 to 2011) are
inconclusive, but could be expected to follow the same trend as the population density metrics
because large RZs in rural areas would likely lessen the effects of increased population density
change.
Recommendations and Conclusion
The ranks of the four largest RZs (i.e., RZ 72, 67, 62, and 8) indicate that RZs are
vulnerable to degradation as a result of smaller habitat patch size, less complex habitat patch
shape, and human-induced disturbances within a habitat patch. All four of the largest RZs have a
substantial proportion of very small wetland habitat patches within them (i.e., habitat < 2 ha).
Because of the high likelihood that these very small wetland habitat patches may be lost in the
future (by definition), and because of the intrinsic difficulty in monitoring their loss, we
recommend that relatively small patches of wetland habitat in the LWRR be a high priority for
future remote-sensing and field-based monitoring and conservation efforts.
Based on the number of human-induced disturbance and patch metric vulnerabilities
among the four largest RZs, wetland habitat in RZ 8 is likely to experience the highest levels of
disturbance and patch fragmentation and/or loss in the future because its vulnerability to all
factors except the unified vulnerability index is greater than the median. The next most
vulnerable RZ among the four largest RZs are: RZ 72 because it has greater than the median
vulnerability to all factors except the unified patch index, population density factors, and the
unified vulnerability index, and RZ 67 because it has greater than the median vulnerability to all
factors except the patch perimeter metric, sinuosity index, circularity index, unified patch index,
and population density factors. The least vulnerable among the four largest RZs is RZ 62
because it has greater than the median vulnerability to all factors except the patch area metric,
patch perimeter metric, sinuosity index, circularity index, unified patch index, unified
vulnerability index, and population density factors. Considering the relative total area of the four
largest RZs (Table 8), the number of small habitat patches (i.e., < 2 ha) in RZs 72, 67, and 62 is
approximately proportional to each refuge's area, leaving RZ 8 with approximately half the
-------
expected number of small habitat patches. RZ 8 (Figure 42; Figure 9) is a unit of the Bald Knob
National Wildlife Refuge, a relatively recent wildlife refuge land acquisition, and is currently
predominated by agricultural land (Figure 9). Thus, RZ 8 is the largest federally owned parcel in
the LWRR to have been recently (and almost completely) impacted by wetland fragmentation,
loss, and human-induced disturbances (Figure 9). Therefore, we recommend that the Bald Knob
National Wildlife Refuge be the highest priority for wetland habitat restoration and protection
among the four largest RZs in the LWRR.
Because indexes are (by definition) derived from other more directly measured metrics
they may be less sensitive to the relative differences among patches than the metrics themselves
(Yoder, 1991; Karr and Chu, 1997). We found that the indexes that included 'patch
characteristics' (i.e., the unified patch index and the unified vulnerability index) tended to rank
the four largest RZs as less vulnerable than their component metrics. That is, the 'patch
characteristic' vulnerability indexes used in this atlas tend to indicate that habitat is less
vulnerable than the component metrics of that index. The purely 'human-induced disturbance'
indexes (i.e., the road index and the unified human index) tended to rank the four largest RZs
consistently with their component metrics, with regard to the median parameter value. Thus, the
results of the unified patch index and unified vulnerability index for RZ 72 (indicating that RZ
72 is relatively less vulnerable than other RZs) may be misleading because of this 'dilution
effect' of combining the sub-component metrics (Table 3a). Accordingly, if index results are
held aside, wetland vulnerability to landscape-ecological degradation factors in RZ 72 and 8 is
similar. Consequently, we recommend that RZ 72, the largest RZ, be given the second highest
priority for wetland restoration and protection among the four largest RZs in the LWRR.
Results of the habitat vulnerability assessment of RZ 67 suggest that it is predominantly
vulnerable to the influence of road construction and the presence of relatively smaller wetland
habitat patches. Human-induced disturbance factors related to the presence of roads in RZ 67
may be partially mitigated by the robustness of patch characteristics within this RZ, with the
exception of patch area. Therefore, we recommend that RZ 67 be given the third highest priority
for continued wetland restoration and protection among the four largest RZs in the LWRR.
Results of habitat vulnerability assessments for RZ 62 suggest that it is primarily
vulnerable to the influence of roads. The human-induced disturbance factors related to the
presence of roads in RZ 62 may be partially mitigated by the robustness of patch characteristics
-------
within this RZ, because all of the habitat size and shape metrics for RZ 62 are relatively high.
Thus, we recommend that RZ 62 be given the lowest priority for continued wetland restoration
and protection among the four largest RZs in the LWRR.
Simulating Landscape Change in the White River National Wildlife Refuge
Because the unified vulnerability index (UVI) integrates habitat patch size, habitat patch
shape, and human-induced disturbance metrics into a single index value it is the most
conservative (i.e., describes the 'least potential impact' scenario) of the measures used to model
habitat vulnerability. That is, a high UVI indicates component metrics that may range from low
vulnerability to extreme vulnerability, but it encompasses all of the subcomponent metrics (see
Table 11 and Table 12; see discussion of using indexes versus component metrics in previous
section). Thus, the UVI was selected to model potential future land cover change in riparian
wetland mallard duck winter habitat, in a portion of RZ 72 (LWRR), given a hypothetical
decrease in the extent and duration of riparian wetland flooding. The hypothetical decrease in
the extent and duration of flooding was assumed to involve a change from "permanently
flooded" or "semi-permanently flooded" wetland conditions to "intermittently flooded" or
"rarely flooded" wetland conditions (after Cowardin et al., 1979; see Table 2a for mallard duck
habitat hydrologic parameters). One hundred-thirteen kilometers of river channel and the
surrounding landscape in the vicinity of the South Unit of the White River National Wildlife
Refuge (Figure 43) was used to simulate these potential future hydrologic changes.
Results of the simulation from an arbitrary upstream point (UTM 15 = 663318E,
3814972N) to an arbitrary downstream terminus (UTM 15 = 674296E, 3769144N) indicate
current wetland habitat vulnerability along the adjacent 226 km of riparian habitat (Figure 44A),
per the UVI. The UVI in patches of mallard duck winter habitat grossly determines the potential
affects of the hypothesized future hydrologic changes (Figure 44B), within wetlands adjacent to
a 30 m region on the riverbank. Mallard duck winter habitat under the simulated future
hydrologic conditions indicates a decrease in the periodicity and duration of flooding along the
226 km of river-adjacent riparian wetlands, which could result in a conversion of twenty-one
percent (2,822 ha) of the 13,514 ha of riparian wetland along the 113 km of White River system
tested (Figure 44B and Figure 44C). Accordingly, seventy-nine percent of the tested wetlands
-------
are relatively less vulnerable to conversion from such decreases in periodicity and duration of
flooding in the future.
The results of this simulation are substantially simplified because they describe solely the
conversion of wetlands from relatively 'wetter conditions' to relatively 'drier conditions'.
Actual wetland change in riparian areas is more complex, involving many biophysical
constraints (Vannote et al., 1980; Gregory et al., 1991; Middleton, 1999). Additionally, because
the UVI is a conservative model, hydrologic changes in this region of the LWRR may result in
substantially greater biological affects, such that facultative-wetland or obligate-wetland plants
(Reed, 1988) might be less able to flourish, resulting in facultative or facultative-upland (Reed,
1988) plant influx and establishment. Thus, future changes in the hydrology of the White River
could result in the loss of plant species that are important resources for wetland organisms (e.g.,
Potamogeton spp. or Polygonum spp. for waterfowl forage). Improved hydrologic models for
the White River could be used to improve upon the assumptions made in this simple example.
Such improvements would help to determine the important linkages between hydrology and the
vulnerability of biological resources for wetland organisms in the Lower White River Region.
-------
ACRONYMS AND ABBREVIATIONS USED
AR-GAP = Arkansas GAP Program
DEM = Digital Elevation Model
EPA = Environmental Protection Agency
ETM+ = Enhanced Thematic Mapper
GAP = GAP Program
GIS = Geographic Information System
HUC = Hydrologic Unit Code
LWRR = Lower White River Region study area
MAVE = Mississippi Alluvial Valley Ecoregion study area
MAVA-LULC = Mississippi Alluvial Valley of Arkansas Landuse/Landcover Project
NAWQA = National Water Quality Assessment Program
NED = National Elevation Dataset
NHD = National Hydrography Dataset
NLCD = National Land Cover Dataset
NWR = National Wildlife Refuge
ORD = Office of Research and Development
PCI = Habitat patch circularity index
PIER = Habitat patch interior to edge ratio
Popdenchg = Habitat patch human population density change from 1990 to 2011
Popdens 1990 = Habitat patch human population density in 1990
Popdens 2011 = Habitat patch human population density in 2011
PRI = Habitat patch road index
PSI = Habitat patch sinuosity index
RARE = Regional Applied Research Effort
Rddens = Habitat patch road density
Rdlen = Habitat patch road length
RZ = Refuge Zone
URL = Universal Reference Locator
USACOE = United States Army Corps of Engineers
U.S. EPA = United States Environmental Protection Agency
U.S. GS = United States Geological Survey
UTM 15 = Universal Transverse Mercator, Zone 15
UVI = Habitat patch unified vulnerability index
WRW = White River Watershed study area
-------
APPENDICES
(CLICK ON LINK)
Appendix A. U.S. Fish and Wildlife Service Publication: Habitat Suitability Index Models:
Mallard (Winter Habitat, Lower Mississippi Valley)
Appendix B. U.S. Fish and Wildlife Service Publication: Habitat Suitability Index Models:
Black Bear, Upper Great Lakes Region
Appendix C. U.S. Fish and Wildlife Service Publication: Habitat Suitability Index Models:
Least Tern
Appendix D. U.S. Fish and Wildlife Service Publication: Classification of Wetlands and
Deepwater Habitats of the United States [Available online at URL:
http://wetl ands. fws. gov/Pub s_Reports/Cl ass_Manual/cl as s_titl epg. htm]
-------
LITERATURE CITED
Allen, A. W. 1987. Habitat Suitability Index Models: Mallard (Winter Habitat, Lower
Mississippi Valley). Biological Report 82(10.132). U.S. Fish and Wildlife Service, Washington,
D.C. 54pp.
American Ornithologists' Union. 1983. Check-list of North American Birds (6th edition).
American Ornithologists' Union, Washington, D.C.
Amstrup, S.C., and J. Beecham. 1976. Activity patterns of radio-collared black bears in Idaho.
Journal of Wildlife Management. 40:340-348.
Anderson, J. R., E.E. Hardy, J.T. Roach, and R.E. Witmer. 1976. A Land Use and Land
Cover Classification System for Use with Remote Sensor Data. U.S. GS Professional Paper 964.
U.S. Geological Survey, Reston, Virginia.
Applied Geographic Solutions. 2001. Census Data, Estimates, and Projections. [Available
online at URL: http://www.appliedgeographic.com/datadescripts.htm.].
Arkansas Game and Fish Commission. 1998. Black Bear Harvest Report, 1997-1998. Arkansas
Game and Fish Commission, Little Rock, Arkansas.
Arkansas State Highway Commission. 2002. Statewide Transportation Improvement Program
for Federal Fiscal Years 2003 through 2005. Dataset. Arkansas State Highway Commission,
Little Rock, Arkansas.
Atwood, J.L., and D.E. Minsky. 1983. Least tern foraging ecology at three major California
breeding colonies. Western Birds. 14:57-72.
Baldassarre, G.A., RJ. Whyte, E.E. Quinlan, and E.G. Bolen. 1983. Dynamics and quality of
waste corn available to postbreeding waterfowl in Texas. Wildlife Society Bulletin. 11:25-31.
-------
Barmore, W.J., and D. Stradley. 1971. Predation by black bear on mature elk. Journal of
Mammalogy. 52:199-202.
Bartonek, J.C., RJ. Blohm, R.K. Brace, F.D. Caswell, K.E. Gamble, H.W. Miller, R.S. Posahala,
and M.P. Smith. 1984. Status and needs of the mallard. Transactions of the North American
Wildlife and Natural Resources Conference. 49:501-518.
Bastin, L., and C.D. Thomas. 1999. The distribution of plant species in urban vegetation
fragments. Landscape Ecology. 14:493-507.
Batema, D.L., G.S. Henderson, and L.H. Frederickson. 1985. Wetland Invertebrate Distribution
in Bottomland Hardwoods as Influenced by Forest Type and Flooding Regime in: Proceedings
of the Fifth Annual Hardwood Conference, University of Illinois, Urbana, pp. 196-202.
Beeman, L.E., and M.R. Pelton. 1980. Seasonal foods and feeding ecology of black bears in the
Smoky Mountains. Bear Biological Association Conference Series. 3:141-148.
Bellrose, F.C. 1976. Ducks, Geese, and Swans of North America. Stackpole Books, Harrisburg,
Pennsylvania. 543pp.
Bent, A.C. 1921. Life Histories of North American Gulls and Terns. U.S. National Museum
Bulletin. 113.
Bosch, W. 1978. A procedure for quantifying certain geomorphical features. Geographical
Analysis. 10:241-247.
Bradford, D.F., S.E. Franson, A.C. Neale, D.T. Heggem, G.R. Miller, and G.E. Canterbury.
1998. Bird species assemblages as indicators of biological integrity in Great Basin rangeland.
Environmental Monitoring and Assessment. 49:1-22.
Bray, H. 1974. Introductory remarks. Proceedings of Eastern Black Bear Workshop. 2:7-10.
-------
Brinson, M.M., H.D. Bradshaw, R.N. Holmes, and J.B. Elkins, Jr. 1980. Litterfall, stemflow,
and throughfall nutrient fluxes in an alluvial swamp forest. Ecology. 61:827-835.
Brinson, M. 1993. A Hydrogeomorphic Classification for Wetlands. WRP-DE-4. U.S. Army
Corps of Engineers Waterways Experiment Station, Vicksburg, Mississippi.
Bromberg, S. M. 1990. Identifying ecological indicators: An environmental monitoring and
assessment program. Journal of the Air Pollution Control Association. 40:976-978.
Brown, S.L. 1984. The role of wetlands in the Green Swamp in: Cypress Swamps (K.C. Ewel,
and H.T. Odum, eds.). University Presses of Florida, Gainesville, Florida.
Brown, M., and JJ. Dinsmore. 1986. Implications of marsh size and isolation for marsh bird
management. Journal of Wildlife Management. 50:392-397.
Bruce, R.G.H., L. Ryszkowski, and M.G. Paoletti. 1993. Landscape Ecology and
Agroecosystems. Lewis Publishers, Boca Raton, Florida.
Burger,!. 1984. Colony stability in least terns. Condor. 86:61-67.
California Least Tern Recovery Team. 1980. California Least Tern Recovery Plan. U.S. Fish
and Wildlife Service (Region 1), Portland, Oregon.
Cardoza, I.E. 1976. The history and status of the black bear in Massachusetts and adjacent New
England states. Massachusetts Division of Fish and Wildlife Research Bulletin 18.
Carpenter, M. 1973. The black bear in Virginia. Virginia Commission of Game and Inland
Fisheries.
Carreker, R.G. 1985. Habitat Suitability Index Models: Least Tern. Biological Report
82(10.103). U.S. Fish and Wildlife Service, Washington, D.C.
-------
Carter, D.L. 1982. Salinity and Plant Productivity in: CRC handbook of Agricultural
Productivity, vol. 1: Plant Productivity (M. Rechcigl, ed.). CRC Press, Boca Raton, Florida.
ppll7-133.
Chambers, W.L. 1908. The present status of the least tern in southern California. Condor.
10:237.
Cobb, S.P. 1992. Studies of the Temporal and Spatial Distribution of Interior Least Tern
Nesting Habitat Along the Lower Mississippi River. U.S. Army Corps of Engineers, Memphis,
Tennessee.
Connell, J.H., and R.O. Slatyer. 1977. Mechanisms of succession in natural communities and
their role in organization. American Naturalist. 111:1119-1144.
Cowan, I. McT. 1972. The status and conservation of bears (Ursidae) of the world -1970.
International Conference of Bear Research and Management. 2:243-367.
Cowardin, L.M., V. Carter, F.C. Golet, and E.T. LaRoe. 1979. Classification of Wetlands and
Deepwater Habitats of the United States. FWS/OBS-79/31. U.S. Fish and Wildlife Service,
Washington, D.C.
Craig, A.M. 1971. Survey of California least tern nesting sites. Special Wildlife Investment
Project, W-54-R-4, Final Report, Job II-5.1. California Department of Fish and Game.
Crumpton, W.G., and J.L. Baker. 1998. Integrating Wetlands into Agricultural Drainage
Systems: Predictions of Nitrate Loading and Loss in Wetlands Receiving Agricultural
Subsurface Drainage. Guest Lecture Series, pp. 121-125. The Ohio State University,
Department of Natural Resources, Columbus, Ohio.
Dahl, T.E. 1990. Wetlands Losses in the United States, 1780s to 1980s. U.S. Fish and Wildlife
Service, Washington, D.C.
-------
Dahl, T.E., and C.E. Johnson. 1991. Status and Trends of Wetlands in the Conterminous United
States, Mid-1970s to Mid-1980s. U.S. Fish and Wildlife Service, Washington, D.C.
Davis, J.C. 1986. Statistics and Data Analysis in Geology (2nd edition). J. Wiley and Sons,
New York, New York.
Davis, G.J., and M.M. Brinson. 1980. Responses of Submersed Vascular Plant Communities to
Environmental Change. FWS/OBS-79/33. U.S. Fish and Wildlife Service, Kearneysville, West
Virginia.
Diamond, J.M. 1974. Colonization of exploded volcanic islands by birds: The supertramp
strategy. Science. 184:803-806.
Day, J.W., Jr., T.J. Butler, and W.G. Conner. 1977. Productivity and nutrient export studies in a
cypress swamp and lake system in Louisiana in: Estuarine Processes, vol. II (M. Wiley, ed.).
Academic Press, New York, New York.
DeLaune, R.D., R.R. Boar, C.W. Lindau, and B.A. Kleiss. 1996. Denitrification in bottomland
hardwood wetland soils of the Cache River. Wetlands. 16:309-320.
Dortch, M.S. 1996. Removal of solids, nitrogen, and phosphorus in the Cache River wetland.
Wetlands. 16:358-365.
Downing, R.L. 1973. A preliminary nesting survey of least terns and black skimmers in the
east. American Birds. 27:946-949.
Downing, R.L. 1980. Survey of interior least tern nesting populations. American Birds.
34:209-211.
Dzwonko, Z., and S. Loster. 1988. Species richness of small woodlands on the western
Carpathian foothills. Vegetatio. 76:15-27.
-------
Ehrenfeld, J.G. 1983. The effects of changes in land-use on swamps of the New Jersey Pine
Barrens. Biological Conservation. 25:253-275.
Ehrenfeld, J.G. and J.P. Schneider. 1991. Chamaecyparis thyoides wetlands and
suburbanization: Effects on hydrology, water quality and plant community composition.
Journal of Applied Ecology. 28:467-490.
Elowe, K.D. 1984. Home Range, Movements, and Habitat Preferences of Black Bear (Ursus
americanus), in Western Massachusetts. M.S. Thesis. University of Massachusetts, Amherst.
Endangered Species Act. 1973. (16USC 1531-1544) Currently Published List oj'Endangered
and Threatened Wildlife and Plants, per 50 CFR 17.11 and 17.12 (December 1999).
Ewel, K.C. 1990. Swamps in: Ecosystems of Florida (R.L. Myers, and J.J. Ewel, eds.).
University of Central Florida Press, Orlando, Florida.
Faanes, C.A. 1983. Aspects of the nesting ecology of least terns and piping plovers in central
Nebraska. Prairie Naturalist. 15:145-154.
Fauth, P.T., E.J. Gustafson, and K.N. Rabenold. 2000. Using landscape metrics to model source
habitat for Neotropical migrants in the Midwestern U.S. Landscape Ecology. 15:621-631.
Fennessy, M.S., and J.K. Cronk. 1997. The effectiveness and restoration potential of riparian
ecotones for the management of nonpoint source pollution, particularly nitrate. Critical Reviews
in Environmental Science and Technology. 27:285-317.
Forman, R.T.T. 1995. Land Mosaics. Cambridge, New York, New York.
Forman, R.T.T., A.E. Galli, and C.F. Leek. 1976. Forest size and avian diversity in New Jersey
woodlots with some land use implications. Oecologia. 26:1-8.
-------
Frame, G.W. 1974. Black bear predation on salmon at Olson Creek, Alaska. Zeitschriftfur
Tierzuchtung und Zuchtungsbiologie (Journal of Animal Breeding and Genetics). 35:23-38.
Fredrickson, L.H. 1979. Floral andFaunal Changes in Lowland Hardwood Forests in Missouri
Resulting from Channelization, Drainage, and Impoundment. FWS/OBS-78/91. U.S. Fish and
Wildlife Service, Washington, D.C.
Frederickson, L.H., andL.S. Taylor. 1982. Management of Seasonally Flooded Impoundments
for Wildlife. Resource Publication 148. U.S. Fish and Wildlife Service, Washington, D.C.
Ganier, A.F. 1930. Breeding of the least tern on the Mississippi River. Wilson Bulletin.
49:134-141.
Garshelis, D.L., and M.R. Pelton. 1980. Activity of black bears in the Great Smoky Mountains
National Park. Journal of Mammalogy. 61:8-19.
Gesch, D., M. Oimoen, S. Greenlee, C. Nelson, M. Steuck, and D. Tyler. 2002. The National
Elevation Dataset. Photogrammetric Engineering & Remote Sensing. 68:5-32.
Giese, L.A., W.M. Aust, C.C. Trettin, and R.K. Kolka. 2000. Spatial and temporal patterns of
carbon storage and species richness in three South Carolina coastal plain riparian forests.
Ecological Engineering. 15:8157-8170.
Gilpin, M.E. 1981. Peninsula diversity pattern. American Naturalist. 118:291-296.
Gorham, B. 1999. Mapping Agricultural Landuse in the Mississippi Alluvial Valley of
Arkansas: Report on the Mississippi Alluvial Valley of Arkansas Landuse/Landcover (MAVA-
LULC) Project. Center for Advanced Spatial Technologies, University of Arkansas,
Fayetteville, Arkansas.
-------
Gosselink, J.G., and R.E. Turner. 1978. The role of hydrology in freshwater wetland
ecosystems in: Freshwater Wetland Ecosystems: Ecological Processes and Management
Potential (R.E. Good, D.F. Whigham, and R.L. Simpson, eds.). Academic Press, New York,
New York.
Gregory, S.V., FJ. Swanson, W.A. McKee, andK.W. Cummins. 1991. An ecosystem
perspective of riparian zones. Bioscience. 41:540-551.
Green, R.H. 1979. Sampling Design and Statistical Methods for Environmental Biologist?,. J.
Wiley and Sons, New York, New York.
Grochfeld, M. 1983. Colony site selection by least terns: physical attributes of sites. Colonial
Waterbirds. 6:205-213.
Hamazaki, T. 1996. Effects of patch shape on the number of organism. Landscape Ecology.
11:299-306.
Hamilton, R.J., and R.L. Marchington. 1980. Denning and related activities of black bears in
the Coastal Plain of North Carolina. Bear Biology Association Conference Series. 3:121-126.
Hamilton, RJ. 1978. Ecology of the Black Bear in Southeastern North Carolina. M.S. Thesis.
University of Georgia, Athens, Georgia.
Hansen, A.J., Risser, P.G., and F. di Castri. 1992. Epilogue: Biodiversity and Ecological Flows
Across Ecotones in: Landscape Boundaries: Consequences for Biotic Diversity and Ecological
Flows (AJ. Hansen, and F. di Castri, eds.). Springer-Verlag, New York, New York.
Hardy, J.W. 1957. The least tern in the Mississippi Valley. Museum of Michigan State
University Biological Series. 1:1-60.
-------
Harger, E.M. 1967. Homing Behavior of Black Bears. Research Report No. 118. Michigan
Department of Conservation.
Harris, L.D. 1984. The Fragmented Forest: Island Biogeography Theory and the Preservation
of Biotic Diversity. University of Chicago Press, Chicago, Illinois.
Hays, M.B. 1980. Breeding Biology of the Least Tern, Sterna albifrons, on the Gulf Coast of
Mississippi. M.S. Thesis. Mississippi State University, Starkville, Mississippi.
Heggem, D.T., A.C. Neale, C.M Edmonds, L.A. Bice, R.D. Van Remortel, and K.B. Jones.
1999. An Ecological Assessment of the Louisiana Tensas River Basin. EPA/600/R-99/016. U.S.
Environmental Protection Agency, Washington, D.C.
Heggem, D.T., C.M. Edmonds, A.C. Neale, L. Bice, and K.B. Jones. 2000. A Landscape
Ecology Assessment of the Tensas River Basin. Environmental Monitoring and Assessment. 64:
41-54.
Heitmeyer, M.E. 1985. Wintering Strategies of Female Mallards Related to Dynamics of
Low land Hardwood Wetlands in the Upper Mississippi Delta. Ph.D. Dissertation. University of
Missouri, Columbia, Missouri.
Heitmeyer, M.E., and L.H. Fredrickson. 1981. Do wetland conditions in the Mississippi delta
hardwoods influence mallard recruitment? Transactions of the North American Wildlife and
Natural Resources Conference. 46:44-47.
Hergert, G.W., S.D. Klausner, D.R. Bouldin, and P.J. Zwerman. 1981. Effects of dairy manure
on phosphorus concentrations and losses in tile effluent. Journal of Environmental Quality.
10:345-349.
-------
Hermy, M., and H. Stieperaere. 1981. An indirect gradient analysis of the ecological
relationships between ancient and recent riverine woodlands to the south of Bruges (Flanders,
Belgium). Vegetatio. 44:43-49.
Herrero, S.M. 1979. Black bears: the grizzly's replacement? in: The Black Bear in Modern
North America (D. Burk, ed.). Boone and Crockett Club/Amwell Press, Clinton, New Jersey.
pp.179-195.
Hey, D.L., M.A. Cardamone, J.H. Sather, and WJ. Mitsch. 1989. Restoration of riverine
wetlands: The Des Plaines River wetlands demonstration project in: Ecological Engineering:
An Introduction to Ecotechnology (WJ. Mitsch, and S.E. Jorgensen, eds.). J. Wiley and Sons,
New York, New York.
Hicks, A.L. 1996. Aquatic Invertebrates and Wetlands: Ecology, Biomonitoring, and
Assessment of Impact from Urbanization, a Literature Review. The Environmental Institute,
Amherst, Massachusetts.
Hopkinson, C.S., Jr., and J.W. Day, Jr. 1980a. Modeling the relationship between development
and storm water and nutrient runoff. Environmental Management. 4:315-324.
Hopkinson, C.S., Jr., and J.W. Day, Jr. 1980b. Modeling hydrology and eutrophication in a
Louisiana swamp forest ecosystem. Environmental Management. 4:325-335.
Hugie, R.D. 1979. Working group report: Central and Northeast Canada and United States in:
The Black Bear in Modern North America (D. Burk, ed.). Boone and Crockett Club/Amwell
Press, Clinton, New Jersey, pp.250-271.
Hunsaker, C. T., and D. E. Carpenter. 1990. Environmental Monitoring and Assessment
Program-EcologicalIndicators. EPA 600/3-90/060. U. S. Environmental Protection Agency,
Research Triangle Park, North Carolina.
-------
Hunsaker, C. T., D.E. Carpenter, and J. Messer. 1990. Ecological indicators for regional
monitoring. Bulletin of the Ecological Society of America. 71:165-172.
Hupp, C.R., and D.E. Bazemore. 1993. Temporal and spatial patterns of wetland sedimentation,
West Tennessee. Journal of Hydrology. 141:179-196.
Hupp, C.R., and E.E. Morris. 1990. A dendrogeomorphic approach to measurement of
sedimentation in a forested wetland, Black Swamp, Arkansas. Wetlands. 10:107-124.
Jehl, D. 2002. Arkansas Rice Farmers Run Dry, and U.S. Remedy Sets Off Debate in: The
New York Times. The New York Times Company, New York, New York. November 11, p.A4.
Jenning, M.D. 1995. Gap analysis today: A confluence of biology, ecology, and geography for
management of biological resources. Wildlife Society Bulletin. 23:658-662.
Jernigan, L.R, J. Parnell, and T. Quay. 1978. Nesting habitats and breeding populations of the
least tern (Sterna albifrons antillarum) in North Carolina. Sea Grant Publication UNC-SG-78-07.
National Oceanic and Atmospheric Administration, Washington, D.C.
Johnston, C. A. 1989. Human impacts to Minnesota wetlands. Journal of the Minnesota
Academy of Science. 55:120-124.
Johnston, C.A. 1994. Cumulative impacts to wetlands. Wetlands. 14:49-55.
Johnston, W.K., F. Ittihadieh, R.M. Daum, and A.F. Pillsbury. 1965. Nitrogen and phosphorus
in tile drainage effluent. Soil Science Society Proceedings. 29:287-289.
Jones, K.B., K.H. Riitters, J.D. Wickham, R.D. Tankersley, Jr., R.V. O'Neill, D.J. Chaloud, E.R.
Smith, and A.C. Neale. 1997. An Ecological Assessment of the United States Mid-Atlantic
Region: A Landscape Atlas. EPA/600/R-97/130. U.S. Environmental Protection Agency,
Washington, D.C.
-------
Jones, K.B., L.R. Williams, A.M. Pitchford, E.T. Slonecker, J.D. Wickham, R.V. O'Neill, D.
Garofalo, and W.G. Kepner. 2000a. A National Assessment of Landscape Change and Impacts
to Aquatic Resources: A 10-Year Strategic Plan for the Landscape Sciences Program.
EPA/600/R-00/001. U.S. Environmental Protection Agency, Washington, D.C.
Jones, K.B., A.C. Neale, M.S. Nash, K.H. Riitters, J.D. Wickham, R.V. O'Neill, and R.D. Van
Remortel. 2000b. Landscape correlates of breeding bird richness across the United States Mid-
Atlantic Region. Environmental Monitoring and Assessment. 63:159-174.
Jones, K.B., D.T. Heggem, T.G. Wade, A.C. Neale, D.W. Ebert, M.S. Nash, M.H. Mehaffey,
K.A. Hermann, A.R. Selle, S. Augustine, LA. Goodman, J. Pedersen, D. Bolgrien, J.M. Viger, D.
Chiang, C.J. Lin, Y. Zhong, J. Baker, and R.D. Van Remortel. 2000c. Assessing landscape
condition relative to water resources in the western United States: A strategic approach.
Environmental Monitoring and Assessment. 64:227-245.
Jones, K.B., A.C. Neale, M.S. Nash, R.D. Van Remortel, J.D. Wickham, K.H. Riitters, and R.V.
O'Neill. 2001. Predicting nutrient and sediment loadings to streams from landscape metrics: A
multiple watershed study from the United States Mid-Atlantic Region. Landscape Ecology.
16:301-312.
Jonkel, C.J., and I.McT. Cowan. 1971. The black bear in the spruce-fir forest. Wildlife
Monographs. 27:1-57.
Jorde, D.G., G.L. Krapu, and R.D. Crawford. 1983. Feeding ecology of mallards wintering in
Nebraska. Journal Wildlife Management. 47:1044-1053.
Jorde, D.G., G.L. Krapu, R.D. Crawford, and M.A. Hay. 1984. Effects of weather on habitat
selection and behavior of mallards wintering in Nebraska. Condor. 86:258-265.
-------
Karr, J.R., andE.W. Chu. 1997. Biological Monitoring and Assessment: Using Multimetric
Indexes Effectively. EPA/235/R97/001. U.S. Environmental Protection Agency, Seattle,
Washington.
Kellman, M. 1996. Redefining roles: plant community reorganization and species preservation
in fragmented systems. Global Ecology and Biogeography Letters. 5:111-116.
Kilgore, K.J., and J.A. Baker. 1996. Patterns of larval fish abundance in a bottomland hardwood
wetland. Wetlands. 16:288-295.
Kitchens, W.M., Jr., J.M. Dean, L.H. Stevenson, and J.M. Cooper. 1975. The Santee Swamp as
a nutrient sink in: Mineral Cycling in Southeastern Ecosystems (F.G. Howell, J.B. Gentry, and
M.H. Smith, eds.). Energy Research and Development Administration Symposium Series
740513. U.S. Government Printing Office, Washington, D.C.
Kleiss, B.A. 1996. Sediment retention in a bottomland hardwood wetland in eastern Arkansas.
Wetlands. 16:321-333.
Klepinger, K.E., and H. Norton. 1983. Black bear in: North American Fur bearers. A
Contemporary Reference (E.F. Deems, Jr., and D. Pursley, eds.). International Association of
Fish and Wildlife Agencies and Maryland Department of Natural Resources, pp. 99-102.
Klett, A. T., and L. M. Kirsch. 1976. Diurnal use of small wetlands by ducks. Journal of
Wildlife Management. 40:351-353.
Krapu, G.L. 1981. The role of nutrient reserves in mallard reproduction. Auk. 98:29-38.
Kress, M.R., M.R. Graves, and S.G. Bourne. 1996. Loss of bottomland hardwood forests and
forested wetlands in the Cache River Basin, Arkansas. Wetlands. 16:258-263.
-------
Landers, J.L., RJ. Hamilton, A.S. Johnson, and R.L. Marchington. 1979. Foods and habitat of
black bears in southeastern North Carolina. Journal of Wildlife Management. 43:143-153.
LeCount, A.L. 1980. Some aspects of black bear ecology in the Arizona chaparral. Bear
Biological Association Conference Series. 3:175-179.
Likens, G.E., F.H. Bormann, N.M. Johnson, D.W. Fisher, and R.S. Pierce. 1970. Effects of
forest cutting and herbicide treatment on nutrient budgets in the Hubbard Brook Watershed-
Ecosystem. Ecological Monographs. 40:23-47.
Lindzey, J.S., W.S. Kordek, G.J. Matula, Jr., and W.P. Piekielek. 1976. The black bear in
Pennsylvania-status movements, values, and management in: Bears - Their Biology and
Management - Proceedings of the 3rd International Conference on Bear Research and
Management (M.R. Pelton, G.E. Folk, and J.W. Lentfer, eds.). pp215-224.
Llewellyn, D.W., G.P. Shaffer, N.J. Craig, L. Creasman, D. Pashley, M. Swan, and C. Brown.
1996. A decision-support system for prioritizing restoration sites on the Mississippi River
alluvial plain. Conservation Biology 10(5): 1446-1455.
Long, K.S., and Nestler, J.M. 1996. Hydroperiod changes as clues to impacts. Wetlands.
16:258-263.
Lopez, R.D., C.B. Davis, and M.S. Fennessy. 2002. Ecological relationships between landscape
change and plant guilds in depressional wetlands. Landscape Ecology. 17:43-56.
Lopez, R.D., and M.S. Fennessy. 2002. Testing the floristic quality assessment index as an
indicator of wetland condition. Ecological Applications. 12:487-497.
Lopez, R.D., andD.T. Heggem. 2001. An Ecological and Habitat Vulnerability Assessment of
the White River Basin. Research Plan EPA/600/R-01/040. U.S. Environmental Protection
Agency, Washington, D.C.
-------
Luoto, M. 2000. Modeling of rare plant species richness by landscape variable in an agricultural
area in Finland. Plant Ecology. 149:157-168.
MacArthur, R., and E.O. Wilson. 1967. The Theory of Island Biogeography. Princeton
University Press, Princeton, New Jersey.
Maehr, D.S., and J.R. Brady. 1984. Food habits of Florida black bears. Journal of Wildlife
Management. 48:230-235.
Massey, B.W. 1971. A Breeding Study of the California Least Tern, 1971. Report 71-9.
California Department of Fish and Game, Wildlife Management Branch Administration.
Massey, B.W. 1974. Breeding biology of the California least tern. Proceedings of the Linnean
Society of New York. 72:1-24.
Massey, B.W., and J.L. Attwood. 1981. Application of Ecological Information to Habitat
Management for the California Least Tern. Program Report 3. U.S. Fish and Wildlife Service,
Laguna Niguel, California.
Massey, B.W., and J.L. Atwood. 1979. Application of Ecological Information to Habitat
Management for the California Least Tern. Program Report 1. U.S. Fish and Wildlife Service,
Laguna Niguel, California.
Massey, B.W., and J.L. Atwood. 1980. Application of Ecological Information to Habitat
Management for the California Least Tern. Program Report 2. U.S. Fish and Wildlife Service,
Laguna Niguel, California.
Massey, B.W., and J.L. Atwood. 1982. Application of Ecological Information to Management
for the California Least Tern. Program Report 4. U.S. Fish and Wildlife Service, Laguna
Niguel, California.
-------
Matula, G.R., Jr. 1974. Behavioral and Physiological Characteristics of Black Bear in
Northeastern Pennsylvania. M.S. Thesis. Pennsylvania State University, University Park.
McArthur, K.L. 1981. Factors contributing to effectiveness of black bear transplants. Journal
of Wildlife Management. 45:102-110.
McDaniel, B., and S. McDaniel. 1963. Feeding of least terns over land. Auk. 80:544.
McDonnell, MJ. 1984. Interactions Between Landscape Elements: Dispersal of Bird
Disseminated Plants in Post-Agricultural Landscapes in: Proceedings of the First International
Seminar on Methodology in Landscape Ecological Research and Planning, Volume 2 (J. Brandt,
and P. Agger, eds.). Roskilde Universitersforlag GoeRuc, Roskilde, Denmark. pp47-59.
McDonnell, M.J., and E.W. Stiles. 1983. The structural complexity of old field vegetation and
the recruitment of bird-dispersed plant species. Oecologia. 56:109-116.
Mclntyre, N.E., and J.A. Wiens. 1999a. How does habitat patch size affect animal movement?
An experiment with darkling beetles. Ecology. 80:2261-2270.
Mclntyre, N.E., and J.A. Wiens. 1999b. Interactions between habitat abundance and
configuration: experimental validation of some predictions from percolation theory. Oikos.
86:129-137.
Meade, R.H. (ed.). 1996. Contaminants in the Mississippi River, 1987-1992. Circular 1133.
U.S. Geological Survey, Reston, Virginia.
Mealey, S.D. 1975. The Natural Food Habits of Free Ranging Grizzly Bears in Yellowstone
National Park, 1973-1974. M.S. Thesis. Montana State University, Bozeman, Montana.
Merian, E. (ed.). 1990. Metals and Their Compounds in the Environment. VCH, Weinheim,
Germany. 1430pp.
-------
Middleton, B.A. 1999. Wetland Restoration, Flood Pulsing and Disturbance Dynamics. John
Wiley and Sons, New York, New York.
Mitsch, W.J., and J.G. Gosselink. 1993. Wetlands (second edition). Van Nostrand Reinhold,
New York, New York.
Moller, T.R., and C.P. Rordam. 1985. Species numbers of vascular plants in relation to area,
isolation, and age of ponds in Denmark. Oikos. 45:8-16.
Moseley, LJ. 1976. Behavior and Communication in the Least Tern (Sterna albifrons). Ph.D.
Dissertation. University of North Carolina, Chapel Hill, North Carolina.
Murkin, H.R. 1986. Relationships between waterfowl and macroinvertebrate densities in a
northern prairie marsh. Journal of Wildlife Management. 50:212-217.
Nagasaka, A., and F. Nakamura. 1999. The influences of land-use changes on hydrology and
riparian environment in a northern Japanese landscape. Landscape Ecology. 14:543-556.
Naugle, D.E., R.R. Johnson, M.E. Estey, and K.F. Higgins. 2000. A landscape approach to
conserving wetland bird habitat in the prairie pothole region of eastern South Dakota. Wetlands.
20:588-604.
Nichols, J.D., K.J. Reinecke, and J.E. Hines. 1983. Factors affecting the distribution of mallards
wintering in the Mississippi Alluvial Valley. Auk. 100:932-946.
NRI (National Resources Inventory). 2000. 1997 National Resources Inventory Wetland Data.
National Resources Conservation Service, Resources Inventory Division, Beltsville, MD.
Oberholser, H.C. 1974. The Bird Life of Texas (Volume 1). University of Texas Press, Austin,
Texas.
-------
Odum, H.T. 1984. Summary: Cypress swamps and their regional role in: Cypress Swamps
(K.C. Ewel, and H.T. Odum, eds.). University Press of Florida, Gainesville, Florida.
Odum, E.P. 1985. Trends expected in stressed ecosystems. Bioscience. 35:419-422.
Ogutu, Z.A. 1996. Multivariate analysis of plant communities in the Narok district, Kenya: The
influence of environmental factors and human disturbance. Vegetatio. 126:181-189.
O'Neill, R.V., C.T. Hunsaker, K.B. Jones, K.H. Riitters, J.D. Wickham, P.M. Schwartz, LA.
Goodman, B.L. Jackson, and W.S. Baillargeon. 1997. Monitoring environmental quality at the
landscape scale. BioScience. 47:513-519.
Omernik, J.M. 1987. Ecoregions of the conterminous United States. Annals of the Association
of American Geographers. 77:188-125.
Opdam, P. 1990. Understanding the ecology of populations in fragmented landscapes in:
Transactions of the 19th International Union of Game Biologists Congress. IUGB, Trondheim,
Norway. pp381-390.
Opdam, P., R.V. Apeldoorn, A. Schotman, and J Kalkhoven. 1993. Population responses to
landscape fragmentation in: Landscape Ecology of a Stressed Environment (C.C. Vos, and P.
Opdam, eds.). Chapman and Hall, London, United Kingdom.
Ozoga, J.J., and L.J. Verme. 1982. Predation by black bears on newborn white-tailed deer.
Journal of Mammalogy. 63:695-696.
Pelton, M.R. 1982. Black bear in: Wild Mammals of North America: Biology, Management,
and Economics (J.A. Chapman, and G.A. Feldhamer, eds.). Johns Hopkins University Press,
Baltimore, Maryland. pp504-514.
-------
Peterjohn, W.T., and D.L. Correll. 1984. Nutrient dynamics in an agricultural watershed:
Observations on the role of a riparian forest. Ecology. 65:1466-1475.
Pickett, S.T.A., and J.N. Thompson. 1978. Patch dynamics and the design of nature reserves.
Biological Conservation. 13:27-37.
Poiani, K.A., B.L. Bedford, and M.D. Merrill. 1996. A GIS-based index for relating landscape
characteristics to potential nitrogen leaching to wetlands. Landscape Ecology. 11:237-255.
Poiani, K. A., and P.M. Dixon. 1995. Seed banks of Carolina bays: Potential contributions from
surrounding landscape vegetation. American Midland Naturalist. 134:140-154.
Presely, B.J., J.H. Trefrey, and R.F. Stokes. 1980. Heavy metal inputs to Mississippi Delta
sediments: A Historical Review. Water, Air, and Soil Pollution. 13:481-494.
Rabalais, N. N., R. E. Turner, Q. Dortch, D. Justic, V. J. Bierman, Jr., and W. J. Wiseman, Jr..
2002. (Review) Nutrient-enhanced productivity in the northern Gulf of Mexico: past, present
and future. Hydrobiologia. 475/476:39-63.
Ramsar (Ramsar Convention Secretariat). 2002. The Ramsar List of Wetlands of International
Importance. Convention on Wetlands, Article 2.1, 1989 amendment. Ramsar Convention,
Ramsar, Iran.
Reed, Jr., P.B. 1988. National List of Plant Species that Occur in Wetlands. Biological Report
88(26.3). U.S. Fish and Wildlife Service, Washington, D.C.
Reynolds, D.G., and J.J. Beecham. 1980. Home range activities and reproduction of black bears
in west-central Idaho. Bear Biological Association Conference Series. 3:181 -190.
-------
Riitters, K.H., R.V. O'Neill, C.T. Hunsaker, J.D. Wickham, D.H. Yankee, S.P. Timmins, K.B.
Jones, and B.L. Jackson. 1995. A factor analysis of landscape pattern and structure metrics.
Landscape Ecology. 10:23-39.
Robinson, A. 1995. Small and seasonal does not mean insignificant: Why it's worth standing up
for tiny and temporary wetlands. Journal of Soil and Water Conservation. Dec:586-590.
Robinson, J.S., and A.N. Sharpley. 1995. Release of nitrogen and phosphorus from poultry
litter. Journal of Environmental Quality. 24:62-67.
Rogers, L.L., D.W. Kuehn, A.W. Erickson, E.M. Harger, L.J. Verme, and J.J. Ozoga. 1976.
Characteristics and management of black bears that feed in garbage dumps, campgrounds, or
residential areas in Bears - Their Biology and Management - Proceedings of the 3rd
International Conference on Bear Research and Management (M.R. Pelton, G.E. Folk, and J.W.
Lentfer, eds.). pp!69-175.
Rogers, L.L. 1976. Effects of mast and berry crop failures on survival, growth, and reproductive
success of black bears. Transactions of the North American Wildlife and Natural Resources
Conference. 41:432-438.
Rogers, L.L. 1987. Effects of food supply and kinship on social behavior, movements, and
population growth of black bears in northeastern Minnesota. Wildlife Monograph 97.
Rogers, L.L., and A.W. Allen. 1987. Habitat Suitability Index Models: Black Bear, Upper
Great Lakes Region. Biological Report 82(10.144). U.S. Fish and Wildlife Service,
Washington, D.C.
Rogers, L.L. 1992. Watchable Wildlife: The Black Bear. U.S. Forest Service, North Central
Forest Experiment Station.
-------
Romkens, M.J., D.W. Nelson, and J.V. Mannering. 1973. Nitrogen and phosphorus
composition of surface runoff as affected by tillage method. Journal of Environmental Quality.
2:292-295.
Roth, N.E., J.D. Allan, and D.L. Erickson. 1996. Landscape influences on stream biotic
integrity assessed at multiple spatial scales. Landscape Ecology. 11:141-156.
Rowan, W. 1928. Bears and bird eggs. Condor. 30:246.
Scott, J.M., F. Davis, B. Csuti, R. Noss, B. Butterfield, C. Groves, H. Anderson, S. Caicco, F.
D'erchia, F., Edwards, Jr., T.C., Ulliman, J., and R.G. Wright. 1993. Gap analysis: a
geographical approach to protection of biological diversity. Wildlife Monographs. 123:1-41.
Shaw, S.P., and C.G. Fredine. 1956. Wetlands of the United States, Their Extent, and Value for
Waterfowl and Other Wildlife. Circular 39. U.S. Fish and Wildlife Service, Washington, D.C.
Schulenberg, E., J., and M. Schulenberg. 1980. Distribution and ecological study of the least
tern in Kansas - Nongame Wildlife Project. Kansas Fish and Game Commission.
Schulenburg, J.H., and M.B. Ptacek. 1984. Status of the interior least tern in Kansas. American
Birds. 36:975-981.
Simberloff, D., and L.G. Abele. 1982. Refuge design and island biogeographic theory: Effects
of fragmentation. The American Naturalist. 120:41-50.
Simberloff, D., and E.O. Wilson. 1970. Experimental zoogeography of islands. A two-year
record of colonization. Ecology. 51:934-937.
Smith, T.R. 1985. Ecology of Black Bears in a Bottomland Hardwood Forest in Arkansas.
Ph.D. Dissertation. University of Tennessee, Knoxville, Tennessee.
-------
Smith, R.D. 1996. Composition, structure, and distribution of woody vegetation on the Cache
River floodplain, Arkansas. Wetlands. 16:264-278.
Smith, K.G., R.S. Dzur, D.G. Catanzaro, M.E. Garner, W.F. Limp, W.F. 1998. The Arkansas
GAP Analysis Project. Final Report [Available online at URL: http://web.cast.uark.edu/gap/].
Soule, M.E., B.A. Wilcox, and C. Holtby, C. 1979. Benign neglect: A model of faunal collapse
in game reserves of East Africa. Biological Conservation. 15:259-272.
Stevens, C.E., A.W. Diamond, and T.S. Gabor. 2002. Anuran call surveys on small wetlands in
Prince Edward Island, Canada restored by dredging of sediments. 22:90-99.
Stiling, P.D. 1992. Ecology: Theories and Applications (Second Edition). Prentice Hall, Upper
Saddle River, New Jersey.
Stoddart, D.R. 1965. The shape of atolls. Marine Geology. 3:369-383.
Strittholt, J.R., and R.E.J. Boerner. 1995. Applying biodiversity gap analysis in a regional
nature reserve design for the edge of Appalachia, Ohio (U.S.A.). Conservation Biology. 9:1492-
1505.
The Nature Conservancy. 1992. The Forested Wetlands of the Mississippi River: An Ecosystem
in Crisis. The Nature Conservancy, Baton Rouge, Louisiana.
Tietje, W.D., B.D. Pelchat, and R.L. Ruff. 1986. Cannibalism of denned black bears. Journal
ofMammology. 67:762-766.
Tomkins, I.R. 1959. Life history notes on the least tern. Wilson Bulletin. 71:313-322.
Triplett, G.B., Jr., and D.M. Van Doren, Jr. 1969. Nitrogen, phosphorus and potassium
fertilization of non-tilled maize. Agronomy Journal. 61:637-639.
-------
Twedt, D.J., and C.R. Loesch. 1999. Forest area and distribution in the Mississippi alluvial
valley: implications fro breeding bird conservation. Journal of Biogeography. 26:1215-1224.
Twedt, D., D. Pashley, C. Hunter, A. Mueller, C. Brown, and B. Ford. 1999. Partners in Flight
Bird Conservation Plan for the Mississippi Alluvial Valley. Public Draft version 1. U. S. Fish
and Wildlife Service, University of Memphis, Memphis, Tennessee.
Unwin, DJ. 1981. Introductory Spatial Analysis. Methuen, London, United Kingdom.
U.S. ACOE (U.S. Army Corps of Engineers). 1999. Biological Assessment: Interior
Population of the Least Tern, Sterna antillarum. U.S. Army Corps of Engineers, Mississippi
Valley Division, Mississippi River Commission, Vicksburg, Mississippi.
U.S. EPA (U.S. Environmental Protection Agency). 1990. Biological Criteria: National
Program Guidance for Surface Waters. EPA-440/5-90-004. U.S. Environmental Protection
Agency, Washington, D.C.
U.S. FWS (U.S. Fish and Wildlife Service). Various. NationalWetland Inventory Dataset. U.S.
Fish and Wildlife Service, Washington, D.C. [Available online at URL:
http: //www. nwi. fws. gov/Map s/map s. htm].
U.S. GS and U.S. EPA (U.S. Geological Survey and U.S. Environmental Protection Agency).
2000. National Hydrography Dataset Standards. U.S. Geological Survey, Reston, Virginia.
[Available online at URL: http://rmmcweb.cr.usgs.gov/public/nmpstds/nhdstds.html].
U.S. GS (U.S. Geological Survey). 1994. Hydrologic Units Maps of the Conterminous United
States (Edition 1). U.S. Geological Survey, Reston, Virginia. [Available online at URL:
http://water.usgs.gov/lookup/getspatial7huc250k].
U.S. GS (U.S. Geological Survey). 1998. The Arkansas GAP Analysis Project. Final Report.
U.S. Geological Survey, Reston, Virginia.
-------
U.S. GS (U.S. Geological Survey). 2001. The Quality of our Nation's Waters. USGS Circular
1225. U.S. Geological Survey, Reston, Virginia. [Available online at URL:
http://water.usgs.gov/pubs/circ/circl225].
Valett, M.H., C.L. Crenshaw, and P.P. Wagner. 2002. Stream and nutrient uptake, forest
succession, and biogeochemical theory. Ecology. 83:2888-2901.
van der Valk, A.G. 1981. Succession in wetlands: A Gleasonian approach. Ecology. 62:688-
696.
van der Valk, A.G., and C.B. Davis. 1980. The impact of a natural drawdown on the growth of
four emergent species in a prairie glacial march. Aquatic Botany. 9:301-322.
Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell, and C.E. Cushing. 1980. The river
continuum concept. Canadian Journal of Fish and Aquatic Science. 37:130-137.
Vogelmann, I.E., S.E. Howard, L. Yang, C.R. Larson, B.K. Wylie, andN. VanDriel. 2001.
Completion of the 1990s National Land Cover Data Set for the Conterminous United States from
Landsat Thematic Mapper Data and Ancillary Data Sources. Photogrammetric Engineering and
Remote Sensing. 67:650-662.
Wakeley, J.S., and T.H. Roberts. 1996. Bird distributions and forest zonation in a bottomland
hardwood wetland. Wetlands. 16:296-308.
Walton, R., I.E. Davis, T.H. Martin, and R.S. Chapman. 1996. Hydrology of the Black Swamp
wetlands on the Cache River, Arkansas. Wetlands. 16:279-287.
Walton, R., R.S. Chapman, and I.E. Davis. 1996. Development and application of the wetlands
dynamic water budget model. Wetlands. 16:347-357.
-------
Warren, LJ. 1981. Contamination of sediments by lead, zinc, and cadmium: A Review.
Environmental Pollution (Series B). 2:401-436.
Whitcomb, R.F. 1977. Island biogeography and "habitat islands" of eastern forest. American
Birds. 31:3-5.
Wickham, J.D., K.B. Jones, K.H. Riitters, R.V. O'Neill, R.D. Tankersley, and E.R. Smith. 1999.
Environmental auditing: An integrated environmental assessment of the U.S. Mid-Atlantic
Region. Environmental Management. 24:553-560.
Wilber, D.H, R.E. Tighe, and LJ. O'Neil. 1996. Associations between changes in agriculture
and hydrology in the Cache River Basin, Arkansas, USA. Wetlands. 16:366-378.
Wilcox, D.A. 1995. Wetland and aquatic macrophytes as indicators of anthropogenic
hydrologic disturbance. Natural Areas Journal. 15:240-248.
Willis, C., and WJ. Mitsch. 1995. Effects of hydrology and nutrients on seedling emergence
and biomass of aquatic macrophytes from natural and artificial seed banks. Ecological
Engineering. 4:65-76.
Wilson, J.B., and W.M. King. 1995. Human-mediated vegetation switches as processes in
landscape ecology. Landscape Ecology. 10:191-196.
Yoder, C. 1991. Answering some concerns about biological criteria based on experiences in
Ohio in: Proceedings of Water Quality Standards for the 21st Century. U.S. Environmental
Protection Agency, Criteria/Standards Division, Washington, D.C. pp.95-104.
Zar, J.H. 1984. Biostatistical Analysis. Prentice Hall, Englewood Cliffs, New Jersey.
-------
Tables
-------
Table la. GIS data sets used to derive mallard duck, black bear, least tern, and wetland plant habitat suitability and habitat vulnerability in the Lower White River Region.
Data Used Derived Land Cover or Land Use Reference or Source Relevant Internet URL
Mississippi Alluvial Valley of Arkansas Agriculture Crop Cover Gorham, 1999 http://www.cast.uark.edu/local/lulc/
Landuse/Landcover (MAVA-LULC)
National Hydrographic Dataset (NHD) v.l Surface Water Location USGS and USEPA, 1999 http://nhd.usgs.gov/
National Wetland Inventory (NWI) Wetland Cover and Hydrology Type U.S. FWS, Various http://www.nwi.fws.gov/
Arkansas GAP Program (AR-GAP) Vegetation Community Cover, Vegetation Taxa Cover, Surface Water Location, Smith etal., 1998 http://web.cast.uark.edu/gap/
Agriculture Cover, Urban Cover
Historical Forest Cover 1950s Forest Cover Llewellyn et al., 1996
U.S. Census Bureau Statistics Current and Future Estimates of Human Population Applied Geographic Solutions, 2001 http://www.appliedgeographic.com/datadescripts.htmtfcens
ussfl
Wessex Streets v.7.0 Road Length and Density Geographic Data Technology, 1999
-------
Table Ib. GIS data sets used to derive mallard ducks, black bears, least terns, and wetland plant habitat suitability and habitat vulnerability in the Mississippi Alluvial Valley Ecoregion.
Data Used Derived Land Cover or Land Use Reference or Source Relevant Internet URL
National Land Cover Dataset (NLCD) Open Water, Residential/Commercial/Industrial, Barren, Agriculture, Vogelmann et al., 2001 http://landcover.usgs.gov/natllandcover.htoil
Forest, Shrubland, Grassland/Herbaceous, Woody Wetland, and Emergent
Herbaceous Wetland Cover
National Hydrographic Dataset (NHD) v. 1 Surface Water Location U.S. GS and U.S. EPA, 1999 http://nhd.usgs.gov/
U.S. Census Bureau Statistics Current and Future Estimates of Human Population Applied Geographic Solutions, 2001 http://www.appliedgeographic.eom/datadescripts.htoi//censussfl
Wessex Streets v.7.0 Road Length and Density Geographic Data Technology, 1999 http://www.geographic.com/home/index.cfm
-------
Table Ic. GIS data sets used to derive water quality vulnerability models in the White River Watershed.
Data Used
Derived Land Cover or Land Use
Reference or Source
Relevant Internet URL
National Land Cover Dataset (NLCD)
National Hydrographic Dataset (NHD) v.l
Wessex Streets v.7.0
National Elevation Dataset (NED)
Open Water, Residential/Conimercial/Industrial, Barren, Agriculture, Forest,
Shrubland, Grassland/Herbaceous, Woody Wetland, and Emergent Herbaceous
Wetland Cover
Surface Water Location
Road Length and Density
Topographic slope
Vogelmann et al., 2001
U.S. GS and U.S. EPA, 1999
Geographic Data Technology, 1999
Gesch et al., 2002
http ://landcover.usgs .gov/natllandcover.html
http ://nhd.usgs .gov/
http://www.geographic.com/home/index.cfm
http://gisdata.usgs.net/NED/default.asp
-------
Table 2a. Data classes used to produce habitat suitability GIS models for mallard ducks in the Lower
White River Region and the Mississippi Alluvial Valley Ecoregion. Hydroperiodicity per Cowardin
etal. (1979).
Lower White River Mississippi Alluvial Valley
Habitat Suitability Class in GIS Models Region Models Ecoregion Models
Wetland with fluctuating hyrdoperiod; trees/shrubs present; oak included; ,
overcup oak excluded
Wetland with fluctuating hyrdoperiod; trees/shrubs present; oak included; /
overcup oak included
Wetland with fluctuating hyrdoperiod; trees/shrubs present; oak excluded V
Wetland with fluctuating hydroperiod; solely herbaceous plants present V
Wetland with fluctuating hydroperiod; no plants present V
Wetland with infrequent flooding or a lake, impoundment, river, or stream; i
trees/shrubs present; oak included; overcup oak excluded
Wetland with infrequent flooding or a lake, impoundment, river, or stream; ,
trees/shrubs present; oak included; overcup oak included
Wetland with infrequent flooding or a lake, impoundment, river, or stream; ,
trees/shrubs present; oak excluded
Wetland with infrequent flooding or a lake, impoundment, river, or stream; /
solely herbaceous plants present
Wetland with infrequent flooding or a lake, impoundment, river, or stream; no ,
plants present
Upland; Agriculture V V
Upland; Non-agriculture V V
Wetland; not open water; solely herbaceous plants present V
Wetland; not open water; woody plants present V
Open water V
-------
Table 2b. Data classes used to produce habitat suitability GIS models for black bears in the Lower White
River Region and the Mississippi Alluvial Valley Ecoregion.
Habitat Suitability Classes used in GIS Models
Wetland; woody plants present; 4 tree species present; forest present in the 1950s
Wetland; woody plants present; 3 tree species present; forest present in the 1950s
Wetland; woody plants present; 2 tree species present; forest present in the 1950s
Wetland; woody plants present; 1 tree species present; forest present in the 1950s
Wetland; woody plants present; 4 tree species present; forest present solely after 1950s
Wetland; woody plants present; 3 tree speciespresent; forest present solely after 1950s
Wetland; woody plants present; 2 tree speciespresent; forest present solely after 1950s
Wetland; woody plants present; 1 tree species present; forest present solely after 1950s
Non- woody wetland; 4 herbaceous plant species present; forest absent in 1950s
Non- woody wetland; 3 herbaceous plant species present; forest absent in 1950s
Non- woody wetland; 2 herbaceous plant species present; forest absent in 1950s
Non- woody wetland; 1 herbaceous plant species present; forest absent in 1950s
Non-woody wetland; no herbaceous plant species present; forest absent in 1950s
Non- woody wetland; 4 herbaceous plant species present; forest present in 1950s
Non- woody wetland; 3 herbaceous plant species present; forest present in 1950s
Non- woody wetland; 2 herbaceous plant species present; forest present in 1950s
Non- woody wetland; 1 herbaceous plant species present; forest present in 1950s
Non-woody wetland; no herbaceous plant species present; forest present in 1950s
Open water
Wetland; trees present, > 250 meters from trees
Wetland; non-agriculture herbaceous plants present; < 250 meters from trees
Wetland; non-agriculture herbaceous plants present; > 250 meters from trees
Upland; trees present; < 250 meters from trees
Upland; non-agriculture herbaceous plants present; < 250 meters from trees
Upland; non-agriculture herbaceous plants present; > 250 meters from trees
Upland, agriculture, urban, or non- vegetated; < 250 meters from trees
Upland, agriculture, urban, or non-vegetated; > 250 meters from trees
White River
Region
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
Valley Ecoregion
Models
V
V
V
V
V
V
V
V
V
-------
Table 2c. Data classes used to produce habitat suitability GIS models for least terns in the Lower White River Region
and the Mississippi Alluvial Valley Ecoregion.
Lower White River Region Mississippi Alluvial Valley
Habitat Suitability Classes used in GIS Models Models Ecoregion Models
Sandy shore present during June-August time period; lotic or lentic ecosystem within 30 meters V
Sandy shore present during June-August time period V
-------
Table 2d. Data classes used to produce habitat suitability GIS models for wetland plants in the Lower White River Region
and the Mississippi Alluvial Valley Ecoregion.
Lower White River Region
Habitat Suitability Classes used in GIS Models Models
Wetland; lacustrine; littoral; rooted vascular plants present; semipermanently flooded V
Wetland; lacustrine; littoral; rooted vascular plants present; permanently flooded V
Wetland; lacustrine; littoral; unconsolidated bottom; permanently flooded V
Wetland; palustrine; aquatic bed; rooted vascular plants present; permanently flooded V
Wetland; palustrine; persistent emergent plants present; temporarily flooded V
Wetland; palustrine; persistent emergent plants present; seasonally flooded v
Wetland; palustrine; persistent emergent plants present; semipermanently flooded V
Wetland; palustrine; broad-leaved deciduous forest present; seasonally flooded V
Wetland; palustrine; broad-leaved deciduous forest present; temporarily flooded V
Wetland; palustrine; broad-leaved deciduous forest present; semipermanently flooded V
Wetland; palustrine; deciduous forest present; temporarily flooded V
Wetland; palustrine; deciduous forest present; seasonally flooded V
Wetland; palustrine; needle-leaved deciduous forest present; seasonally flooded V
Wetland; palustrine; needle-leaved deciduous forest present; semipermanently flooded v
Wetland; palustrine; broad-leaved deciduous shrubs present; seasonally flooded V
Wetland; palustrine; broad-leaved deciduous shrubs present; temporarily flooded V
Wetland; palustrine; broad-leaved deciduous shrubs present; semipermanently flooded V
Wetland; palustrine; unconsolidated bottom V
Wetland; palustrine; unconsolidated shore; seasonally flooded V
Wetland; riverine; lower perennial; unconsolidated bottom; permanently flooded V
Wetland; riverine; lower perennial; unconsolidated bottom; temporarily flooded V
Wetland; riverine; lower perennial; unconsolidated bottom; seasonally flooded v
Upland V
Wetland; woody plants present
Wetland; soley herbaceous plants present
Open water
-------
Table 3a. Data classes used to produce habitat vulnerability G1S models for mallard ducks, black bears, least terns,
and wetland plants in the Lower White River Region and the Mississippi Alluvial Valley Ecoregion.
Habitat Vulnerability Parameter
Interpretation and calculation
Wetland habitat patch total area for patches greater than
2 ha
Smaller patches are relatively less likely to rebound from disturbances (i.e., are more likely to he fragmented or destroyed after
changes in hydrology, destruction of vegetation, or the establishment of opportunistic flora and fauna) than larger patches.
Wetland habitat patch total perimeter length for patches
greater than 2 ha
Patches with shorter perimeters are relatively less likely to rebound from disturbances than patches with longer perimeters.
That is, shorter perimeter length values indicate that a patch has a greater likelihood of fragmentation or loss as a result of
environmental change, such as changes in hydrology, destruction of vegetation, or the establishment of opportunistic flora and
fauna.
Wetland habitat patch interior to edge ratio for patches
greater than 2 ha
Patches with a smaller interior to edge ratio are relatively less likely to rebound from disturbances than patches with a larger
interior to edge ratio. That is, smaller ratio values indicate that a patch has a greater likelihood of fragmentation or loss as a
result of environmental change, such as changes in hydrology, destruction of vegetation, or the establishment of opportunistic
flora and fauna. Calculation: [Area| /1Perimeter], where Perimeter = the patch perimeter and Area = the patch area.
Wetland habitat patch sinuosity index for patches greater
than 2 ha
Patches with a smaller sinuosity index are less winding or convoluted in shape, thus are relatively less likely to rebound from
disturbances than patches with a larger sinuosity index. That is, smaller index values indicate that a patch has a greater
likelihood of fragmentation or loss as a result of environmental change, such as changes in hydrology, destruction of vegetalior
or the establishment of opportunistic flora and fauna. Calculation: (Perimeter] / |2 * it * [v'fArea / JT)]}, where Perimeter = the
patch perimeter and Area = the patch area (after Bosch, 1978 and Davis. 1986).
Wetland habitat patch circularity index for patches
greater than 2 ha
Patches with a smaller circularity index are more circle-like in shape, thus are relatively less likely to rebound from
disturbances than patches with a larger circularity index. That is, smaller index values indicate that a patch has a greater
likelihood of fragmentation or loss as a result of environmental change, such as changes in hydrology, destruction of vegetalior
or the establishment of opportunistic flora and fauna. Calculation: fir * (Perimeter' (2 * ,T)|2} / |Arca|, where Perimeter = the
patch perimeter and Area = the patch area (after Sloddart, 1965 and Unwin, 1981).
Wetland habitat unified patch index for patches greater
than 2 ha
Patches with a smaller unified patch index are relatively less likely to rebound from disturbances than patches with a larger
Unified Patch Index. That is, smaller index values indicate that a patch has a greater likelihood of fragmentation or loss as a
result of environmental change, such as changes in hydrology, destruction of vegetation, or the establishment of opportunistic
flora and fauna. Calculation: Pier * CIRC_ ind * SINHnd, where Pier = the patch interior to edge ratio, ClRCJnd = the patch
Circularity Index, and SIN ind = the patch Sinuosity Index.
-------
Habitat Vulnerability Parameter
Interpretation and calculation
Wetland habitat patch total road length for patches
greater than 2 ha
Patches with greater total road length are relatively less likely to rebound from disturbances than patches with a lesser total road
length. That is, greater length values indicate that a patch has a greater likelihood of fragmentation or loss as a result of
environmental change, such as changes in hydrology, destruction of vegetation, or the establishment of opportunistic flora and fauna.
The increased presence of roads may also bring about the aforementioned disturbances.
Wetland habitat patch total road density for patches
greater than 2 ha
Patches with greater total road density are relatively less likely to rebound from disturbances than patches with a lesser total road
density. That is, greater density values indicate that a patch has a greater likelihood of fragmentation or loss as a result of
environmental change, such as changes in hydrology, destruction of vegetation, or the establishment of opportunistic flora and fauna.
The increased presence of roads may also bring about the aforementioned disturbances.
Wetland habitat patch road index for patches greater
than 2 ha
Patches with a greater road index are relatively less likely to rebound from disturbances than patches with a lesser Road Index. That
is, greater index values indicate that a patch has a greater likelihood of fragmentation or loss as a result of environmental change, such
as changes in hydrology, destruction of vegetation, or the establishment of opportunistic flora and fauna. The increased presence of
roads may also bring about the aforementioned disturbances. Calculation: Rddens + Rdlen, where Rddens = the number of roads per
patch area and Rdlen is the total length of road per patch area.
Wetland habitat patch human population density in
1990 for patches greater than 2 ha
Patches that reside in census block groups with a greater population density are relatively less likely to rebound from disturbances
than patches that reside in areas of lesser population density. That is, greater density values indicate that a patch has a greater
likelihood of fragmentation or loss as a result of environmental change, such as changes in hydrology, destruction of vegetation, or
the establishment of opportunistic flora and fauna. The increased presence of residents near patches may also bring about the
aforementioned disturbances.
Wetland habitat patch human population density in
2011 for patches greater than 2 ha
Patches that reside in census block groups with a greater population density in the future are relatively less likely to rebound from
disturbances than patches that reside in areas of lesser population density in the future. That is, greater density values indicate that a
patch has a greater likelihood of fragmentation or loss as a result of environmental change, such as changes in hydrology, destruction
of vegetation, or the establishment of opportunistic flora and fauna. The increased presence of residents near patches may also bring
about the aforementioned disturbances.
Wetland habitat patch human population density
change from 1990 to 2011 for patches greater than 2 ha
Patches that reside in census tracts with a greater increase in population density are relatively less likely to rebound from disturbances
than patches that reside in areas of lesser population density change. That is, greater (predicted) increases in human population
density indicate that a patch has a greater likelihood of fragmentation or loss as a result of environmental change, such as changes in
hydrology, destruction of vegetation, or the establishment of opportunistic flora and fauna. The increased density of human
population near patches may also bring about the aforementioned disturbances. Calculation: popden2-popdenl, where popden2 =
2011 population density and popdenl = population density in 1990.
Wetland habitat patch unified human index for patches
greater than 2 ha
Patches with a greater unified human index are relatively less likely to rebound from disturbances than patches with a smaller unified
human index. That is, greater index values indicate that a patch has a greater likelihood of fragmentation or loss as a result of
environmental change, such as changes in hydrology, destruction of vegetation, or the establishment of opportunistic flora and fauna.
The increased presence of roads and growth of residential areas near patches may also bring about the aforementioned disturbances.
Calculation: [(popdenchg90_l 1) + C] + [V (rdden+len)], where popdenchg90_l 1 = the population density change between 1990 and
2011, C = a normalization quantity to ensure that the net change value is positive, and rdden+len = the Road Index.
Wetland habitat patch unified vulnerability index for
patches greater than 2 ha
Patches with a smaller unified vulnerability index are less likely to rebound from disturbances than patches with a larger unified
vulnerability index. That is, smaller index values indicate that a patch has a greater likelihood of fragmentation or loss as a result of
environmental change, such as changes in hydrology, destruction of vegetation, or the establishment of opportunistic flora and fauna.
Calculation: [1 / (VUPI)] + (V UHI), where UPI = the patch Unified Patch Index and Uffl = the patch Unified Human Index.
Table 3a (continued)
-------
Table 3b. Metrics used to produce water quality vulnerability GIS models in the
Lower White River Region, measured among 8-digit Hydrologic Unit Codes (HUCs).
"Natural land cover" is a combination of National Land Cover Dataset codes 31,41,
42, 43, 51, 71, 91, and 92 in Table 4).
Water Quality Vulnerability Metric (compared among HUCs)
Largest forest patch proportion of HUC
Mean forest patch area
Largest forest patch area
Forest patch number
Forest patch density
Percent streams within 30 meters of roads
Percent total agriculture on slopes greater than 3%
Percent total agriculture within entire HUC
Percent crop agriculture within entire HUC
Percent pasture within entire HUC
Percent forest within entire HUC
Percent wetland within entire HUC
Percent natural land cover within entire HUC
Percent total agriculture within a 300 meter riparian zones at 30 meter increments
Percent crop agriculture within a 300 meter riparian zones at 30 meter increments
Percent pasture within a 300 meter riparian zones at 30 meter increments
Percent forest within a 300 meter riparian zones at 30 meter increments
Percent wetland within a 300 meter riparian zones at 30 meter increments
Percent natural land cover within a 300 meter riparian zones at 30 meter increments
-------
Table 4. National Land Cover Dataset (NLCD) categories. NLCD
codes were used to develop land cover metrics in this atlas. See
Table Ic for more information about this data set.
Land Cover Category NLCD Land Cover Code
Open Water 11
Perennial Ice/Snow 12
Low Intensity Residential 21
High Intensity Residential 22
Commercial/Industrial/Transportation 23
Bare Rock/Sand/Clay 31
Quarries/Strip Mines/Gravel Pits 32
Transitional 33
Deciduous Forest 41
Evergreen Forest 42
Mixed Forest 43
Deciduous Shrubland 51
Evergreen Shrubland 52
Mixed Shrubland 53
Orchards/Vineyards/Other 61
Grasslands/Herbaceous 71
Pasture/Hay 81
Row Crops 82
Small Grains 83
Fallow 84
Urban/Recreational Grasses 85
Woody Wetlands 91
Emergent Herbaceous Wetlands 92
-------
Table 5. Spearman Rank Correlation of surface water parameters and
percent agriculture land cover within 8-digit HUC subwatersheds in the
White River Watershed. Correlation (Rho) values shown, P < 0.0001
Water Quality Parameter
Dissolved organic carbon
Amino and organic nitrogen
Total phosphorus
Suspended sediment
N
344
188
367
424
Watershed Percent
Total Agriculture
0.697
0.510
0.707
0.692
Streamside Percent
Total Agriculture
0.651
0.532
0.682
0.643
-------
Table 6. Spearman Rank Correlation of surface water parameters and forest land cover metrics within 8-digit HUC
subwatersheds in the White River Watershed. Correlation (Rho) values shown, P < 0.0001
Water Quality Parameter
Dissolved organic carbon
Amino and organic nitrogen
Total phosphorus
Suspended sediment
N
344
188
367
424
Largest forest patch
proportion of HUC
-0.548
-0.202
-0.495
-0.496
Mean forest
patch area
-0.716
-0.507
-0.706
-0.713
Forest Metric
Largest forest patch
area
-0.722
-0.480
-0.624
-0.687
Watershed Percent
Forest
-0.693
-0.570
-0.709
-0.692
Streamside Percent
Forest
-0.693
-0.569
-0.706
-0.692
-------
Table 7. Area of potential mallard duck winter habitat in the Lower White River Region.
Mallard Duck Winter Habitat Suitability Class Habitat Area (ha)
Wetland with fluctuating hyrdoperiod; trees/shrubs present; oak included; overcup oak excluded 82,164
Wetland with fluctuating hyrdoperiod; trees/shrubs present; oak included; overcup oak included 21,173
Wetland with fluctuating hyrdoperiod; trees/shrubs present; oak excluded 77,036
Wetland with fluctuating hydroperiod; solely herbaceous plants present 8,564
Wetland with fluctuating hydroperiod; no plants present 5,730
Wetland with infrequent flooding or a lake, impoundment, river, or stream; trees/shrubs present; oak included; overcup oak excluded 1,924
Wetland with infrequent flooding or a lake, impoundment, river, or stream; trees/shrubs present; oak included; overcup oak included 486
Wetland with infrequent flooding or a lake, impoundment, river, or stream; trees/shrubs present; oak excluded 7,344
Wetland with infrequent flooding or a lake, impoundment, river, or stream; solely herbaceous plants present 2,772
Wetland with infrequent flooding or a lake, impoundment, river, or stream; no plants present 19,875
Upland; Agriculture 423,763
Upland; Non-agriculture 241,246
Total 892,077
-------
Table 8. Area and perimeter of each of the 72 federal
refuge zones in the Lower White River Region, in
ascending order of refuge zone area. Note: four
largest (area) federal refuge zones (8, 62, 67, and 72)
are listed in red type.
Federal Refuge Zone ID
12
64
2
22
53
9
59
26
25
41
31
28
24
47
51
55
57
52
14
13
19
33
17
35
54
3
16
50
37
34
21
1
11
38
42
45
40
44
63
66
32
27
65
60
68
36
39
71
10
18
30
5
46
61
69
6
23
49
48
70
7
20
4
58
43
15
29
56
67
62
8
72
Sum
Mean
1 S.D.
Area (Ha)
7.6
7.7
8.0
8.4
8.4
15.0
15.1
15.9
16.0
16.2
16.3
16.9
17.7
19.6
22.3
22.8
28.1
31.3
33.7
34.0
35.3
39.9
40.0
40.1
40.3
56.3
65.0
65.6
66.5
67.7
78.5
79.4
80.8
81.3
89.4
92.3
95.1
96.4
99.5
106.4
114.2
124.4
128.7
131.4
161.1
167.3
174.3
205.2
212.4
213.5
230.7
250.9
251.4
287.3
295.1
372.8
401.8
407.6
441.6
510.3
738.1
740.6
757.3
767.9
1094.4
1261.4
1629.0
1756.1
2622.5
2899.8
5325.8
64552.1
89529.1
2558.0
10838.3
Perimeter (m)
1182
1237
1208
1211
1204
1548
2427
1596
1600
2025
1617
1646
1684
2112
1895
1899
2702
2353
2485
2475
2514
3799
2808
3823
2970
3132
4024
3773
3263
5196
4815
7156
3847
4032
5606
5224
4971
4822
5824
4380
4459
8113
4811
6252
7609
9596
10229
8099
7158
8143
10935
14060
9844
10260
11304
13889
10756
9278
9046
13622
25561
18248
25088
19509
30733
30217
31502
32308
42974
44889
52555
335315
891387
25468
55365
-------
Table 9. Summary table of Spearman Rank Correlation between federal refuge zone area and: (a) area of habitat patches,
(b) percent contribution of habitat patches, (c) the original area of the habitat patches (inside and outside of the zone), (d) habitat
patch perimeter, (e) habitat patch interior:edge ratio, (f) habitat patch sinuosity index, (g) habitat patch circularity index, (h) unified
patch index, (i) habitat patch road density, (j) habitat patch road length, (k) habitat patch road index, (1) population density in
the year 1990, (m) estimated population density in the year 2011, (n) estimated population density change from 1990 to 2001,
(o) unified human index, and (p) unified vulnerability index within a zone. Correlation (Rho) values and significance are shown
for all refuge zones (N = 72), and for all refuge zones except the four largest (N = 68, i.e., excluding refuge zones 8, 62, and 67,
and 72); ns = not significant, * = 0.05, ** = 0.01, *** = 0.001, **** = 0.0001. All calculations are based upon habitat patches > 2 ha.
Habitat Patch Metric or Index Name
Area of All Federal Refuge Area of All Federal Refuge Zones Except
Zones Four Largest Refuge Zones
Area of habitat patches within a refuge zone
Percent contribution of habitat patches within a refuge zone
Original (i.e., inside and outside) area of the habitat patches for a refuge zone
Habitat patch perimeter within a refuge zone
Habitat patch interior:edge ratio within a refuge zone
Habitat patch sinuosity index within a refuge zone
Habitat patch circularity index within a refuge zone
Unified patch index within a refuge zone
Habitat patch road density within a refuge zone
Habitat patch road length within a refuge zone
Habitat patch road index within a refuge zone
Population density in the year 1990 within a refuge zone
Estimated population density in the year 2011 within a refuge zone
Estimated population density change from 1990 to 2001 within a refuge zone
Unified human index within a refuge zone
Unified vulnerability index within a refuge zone
0.127****
-0 723 ****
-0.141 ****
-0.095 ****
-0.188****
ns
0.032****
-0 214 ****
-0.233 ****
-0.039 *
0.116**
-0.503 ****
-0.149****
-0.125 ***
-0.148****
ns
-0.026 ****
-0.113 **
-0.114**
ns
-------
Table 10. Rank of each of the 72 federal refuge zones in the Lower White River Region by (a) zone area, (b) zone perimeter,
(c) mean area of habitat patches within a zone, and (d) mean percent contribution of habitat patches within a zone. Federal
refuge zone ranks are in ascending order for each parameter. All calculations are based upon habitat patches > 2 ha.
Note: four largest (area) federal refuge zones (8, 62, 67, and 72) are listed in red type.
Rank of Federal Wildlife Refuge Zone Metric (Zones 1-72)
Federal Refuge Zone
Area (ha)
12
64
2
22
53
9
59
26
25
41
31
28
24
47
51
55
57
52
14
13
19
33
17
35
54
3
16
50
37
34
21
1
11
38
42
45
40
44
63
66
32
27
65
60
68
36
39
71
10
18
30
5
46
61
69
6
23
49
48
70
7
20
4
58
43
15
29
56
67
62
8
72
Federal Refuge Zone
Perimeter (km)
12
53
2
22
64
9
26
25
31
28
24
51
55
41
47
52
59
13
14
19
57
17
54
3
37
50
33
35
11
16
38
66
32
65
21
44
40
34
45
42
63
60
1
10
68
71
27
18
48
49
36
46
39
61
23
30
69
70
6
5
20
58
4
7
15
43
29
56
67
62
8
72
Mean Original Habitat Patch Area Within
Federal Refuge Zone
12
24
25
2
53
44
29
64
40
22
50
17
9
59
19
42
33
26
34
14
18
16
28
47
48
60
51
3
54
55
30
31
71
8
6
61
39
4
35
45
5
7
21
27
15
41
13
69
10
1
46
65
37
66
58
23
36
63
68
43
67
57
20
56
38
52
11
72
62
32
70
49
Mean Percent Contribution of Each Habitat Patch to Relative Metric Value
Federal Refuge Zone Guide
72 Low
29
8
67
15
62
4
56
48
7
43
58
44
6
18
40
20
24
61
30
12
23
69
5
71
42
50
60
46
39
10
34
25
17
70
27
16
65
36
33
45
19
68
3
66
21
49
14
1
54
63
37
2
9
53
35
59
26
51
47
38
32
64
22
55
13
11
28
31
41
57
52 High
-------
Table 11. Rank of each of the 72 federal refuge zones in the Lower White River Region by mallard duck winter habitat patch
characteristics: (a) habitat patch area, habitat patch perimeter, [c] habitat patch interioredge ratio within a zone (PIER),
[d] habitat patch sinuosity index within a zone (PSI), [e] habitat patch circularity index within a zone (PCI), and [f] unified
patch index within a zone (UPI). The gradients of parameter values and vulnerability among the 72 federal refuge zones are
indicated from 'High'to'Low'. A double line indicates median. All calculations are based upon habitat patches > 2 hectares.
Note: four largest (area) federal refuge zones (8, 62, 67, and 72) are listed in red type.
Rank of Federal Wildlife Refuge Zone Metric or Index (Zones 1-72)
Patch Area (ha)
24
50
40
59
53
60
18
51
54
8
47
44
9
6
37
30
65
4
17
15
19
26
34
3
7
71
5
2
16
1
48
10
42
67
72
14
23
43
57
58
36
33
62
39
68
55
38
46
21
45
69
22
61
31
52
56
11
70
66
29
35
13
63
28
32
12
64
27
25
20
41
49
Patch Perimeter (km)
24
40
59
60
53
50
54
37
51
18
9
8
65
6
30
47
2
4
5
3
15
44
48
7
19
34
17
33
42
1
10
72
43
58
26
71
52
14
67
23
57
16
62
36
39
55
56
38
31
21
69
68
46
22
29
20
11
61
45
70
13
12
27
28
66
25
63
35
32
41
64
49
PIER
50
24
44
59
62
47
72
8
71
20
46
53
6
67
48
40
15
26
60
4
30
69
7
17
58
45
19
43
65
18
34
1
39
36
35
10
37
9
51
5
56
21
23
61
68
70
16
66
14
3
64
22
33
11
25
54
63
42
38
2
28
55
27
57
49
29
13
32
41
31
12
52
PSI
54
40
52
24
60
53
2
59
18
9
5
3
42
1
4
30
10
6
51
33
43
15
14
65
8
7
37
36
19
23
72
57
17
34
58
48
56
62
39
31
44
67
47
20
50
38
29
69
55
16
71
26
46
21
68
11
61
22
45
49
27
12
70
13
25
66
63
28
32
41
64
35
PCI
54
52
40
24
53
60
2
59
9
18
3
42
5
51
30
33
4
10
6
1
65
37
15
8
57
7
43
14
34
19
31
48
17
23
47
72
56
36
44
50
58
39
38
62
67
16
20
55
26
29
21
71
69
11
22
68
46
12
13
27
61
25
70
49
45
28
41
66
63
32
35
64
Relative Metric or Index Relative Vulnerability to
UPI Value Guide Disturbance Guide
24 Low High
40
54
60
59
53
37
50
51
9
18
2
65
3
8
47
52
6
5
30
42
34
33
48
4
7
15
19
17
44
57
10
26
1
16
43
14
58
31
23
38
71
55
39
36
67
21
72
56
62
22
69
29
11
68
13
46
12
20
70
28
45
27
61
25
41
66
63
32
35
49
64 High Low
-------
Table 12. Rank of each of the 72 federal refuge zones in the Lower White River Region by mallard duck winter habitat patch
human-induced disturbance characteristics: [a] habitat patch road density within a zone (Rddens), [b] habitat patch road length
within a zone (Rdlen), [c] habitat patch road index within a zone (PRI), [d] population density in the year 1990 within a zone
(Popdensl990), [e] estimated population density in the year 2011 within a zone (Popdens2011), [f] estimated population
density change from 1990 to 2001 in a zone (Pdenchg 1990-2011), [g] unified human index within a zone (UHI), and [h] unified
vulnerability index within a zone (UVI). The gradients of parameter values and vulnerability among the 72 federal refuge zones
are indicated from 'High' to 'Low'. A double line indicates median. All calculations are based upon habitat patches > 2 ha.
Note: four largest (area) federal refuge zones (8, 62, 67, and 72) are listed in red type.
Rank of Federal Wildlife Refuge Zone Metric or Index (Zones 1-72)
Rddens
37
65
56
3
19
31
69
13
72
62
49
12
10
40
8
71
34
64
58
30
39
15
48
67
1
11
41
63
18
70
46
23
17
43
21
29
20
14
4
27
66
45
6
61
7
25
32
35
28
22
68
38
55
36
16
57
52
26
42
33
44
5
2
47
9
51
54
50
53
60
59
24
Rdlen
49
20
12
56
13
31
37
64
58
3
70
41
11
63
1
72
69
10
19
67
62
21
66
34
71
8
27
14
45
25
61
39
65
30
23
40
15
46
17
29
43
7
48
18
6
4
32
35
28
22
68
38
55
36
16
57
52
26
42
33
44
5
2
47
9
51
54
50
53
60
59
24
PRI
49
20
12
56
13
31
37
64
58
3
70
41
11
63
1
72
69
10
19
67
62
21
66
34
71
8
27
14
45
25
61
65
39
30
23
40
15
46
17
29
43
48
7
18
6
4
35
32
28
68
22
36
55
38
16
26
57
44
33
42
5
52
47
2
9
51
50
53
59
60
54
24
Popdensl990
66
12
63
8
64
61
13
41
43
42
57
40
31
38
49
45
44
62
2
47
70
52
1
3
71
27
58
67
46
20
5
39
30
68
60
28
55
51
36
32
56
65
34
4
26
16
69
35
33
72
6
29
37
23
25
11
17
22
10
7
15
19
21
48
14
18
9
54
50
53
59
24
Popdens2011
66
1
2
8
12
64
61
57
41
13
63
43
42
20
45
38
62
49
47
31
40
70
52
71
44
55
58
3
67
46
35
68
32
60
28
51
39
56
65
16
5
69
27
33
72
30
34
6
36
11
17
29
25
4
37
10
7
23
15
22
48
14
21
26
19
18
9
54
50
53
59
24
Pdenchg
1990-2011
1
2
66
8
20
57
64
35
45
55
32
56
61
46
41
11
10
71
14
65
17
48
7
18
28
68
16
52
47
9
51
50
53
59
60
54
24
58
69
15
70
21
62
33
25
67
39
22
19
72
6
13
29
23
49
37
42
5
12
38
34
3
36
4
43
30
44
27
26
31
40
63
um
49
12
13
1
64
37
31
2
41
20
3
66
56
11
8
70
19
58
25
69
21
10
71
45
57
63
72
27
14
62
39
67
65
34
61
23
29
15
46
17
18
48
35
30
55
7
32
6
28
68
16
52
47
9
51
50
53
59
60
54
24
33
40
22
43
42
5
4
38
36
44
26
Relative Metric or Index
UVI Value Guide
12 High
41
57
28
52
50
49
20
56
31
63
66
1
58
64
70
37
11
8
72
69
3
10
40
67
62
21
19
27
71
14
45
61
25
65
34
2
39
30
43
46
15
23
17
29
4
13
48
7
26
18
44
6
5
36
55
38
35
42
22
32
33
68
16
47
9
51
53
59
60
54
24 Low
Relative Vulnerability to
Disturbance Guide
High
Low
-------
Table 13. Assessment of the presence of wetlands less than or equal to 2 ha within each of
68 federal refuge zones in the Lower White River Region. Four federal refuge zones (2, 22, 28,
and 31) did not contain habitat patches less than 2 ha. Federal refuge zones are ranked by
ascending areal contribution of wetlands less than 2 ha within the zone, and include (a) areal
percent contribution, (b) mean patch area (+/- S.D.), and (c) total number of habitat patches
less than 2 ha in federal refuge zone. All calculations are based upon habitat patches less than
2 ha. Note: four largest (area) federal refuge zones (8, 62, 67, and 72) are listed in red type.
Federal Refuge Zone
ID
72
8
62
67
56
29
20
15
43
23
7
58
48
4
54
37
45
11
32
60
63
68
71
46
61
18
36
69
10
30
70
6
5
49
40
66
39
27
55
19
35
16
42
44
65
14
50
38
34
24
59
13
17
21
1
57
33
64
47
51
12
52
3
41
26
25
53
9
Sum
Mean
1 S.D.
Mean Percent Contribution
to Federal Zone Area (ha)
0.000004
0.000058
0.000081
0.000098
0.000120
0.000144
0.000168
0.000235
0.000258
0.000349
0.000380
0.000411
0.000423
0.000435
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.001000
0.002000
0.002000
0.002000
0.002000
0.003000
0.003000
0.003000
0.003000
0.003000
0.003000
0.003000
0.004000
0.004000
0.004000
0.004000
0.005000
0.005000
0.005000
0.006000
0.006000
0.006000
0.007000
0.007000
0.008000
0.009000
0.011000
0.012000
0.015000
0.015000
0.016000
0.018000
0.019000
0.021000
0.023000
0.282164
0.004149
0.005525
Mean Habitat Patch
Area (ha)
2814
3110
2356
2563
2107
2343
1244
2961
2825
1400
2801
3157
1869
3293
508
817
931
1080
1261
1415
1450
1462
1509
1747
1845
1880
2082
2139
2400
2600
2941
2950
3279
3460
1664
2163
2769
2931
640
900
1259
2006
2319
2696
3707
1319
2482
2879
2893
805
808
1572
2399
4487
4519
1842
2645
649
1684
2537
900
4656
8593
2590
2885
3022
1801
3459
157082
2310
1228
Std. Dev. (ha)
3550
3944
3157
3291
2741
2915
1816
3773
3306
1600
3502
4212
2661
4089
795
306
1299
380
2154
2109
1755
1915
2152
1554
2513
2460
rrn
2544
2578
2926
3859
3511
3627
4664
2289
2965
4365
4334
247
-
1848
1815
1765
3853
3967
3597
5049
3636
862
986
1958
2640
5975
5104
2133
3572
823
1108
2922
0
5776
2986
2594
2329
2695
174630
2729
1323
Total Habitat Patch
Count
11391
510
460
631
552
199
184
281
182
21
237
151
76
156
18
4
17
10
33
72
36
63
154
99
75
52
33
127
64
111
36
153
86
29
27
18
37
20
5
1
28
7
4
39
48
1
11
23
7
3
13
8
8
7
12
12
26
4
2
3
2
4
1
3
3
5
1
5
16701
246
1378
-------
Figures
-------
SPRINGFIELD, MO
Lake Taneycomo
Table Rock Lake jj Bull Shoals Lake
FAYETTEVILLE, AR
FORT SMITH, OK -
PINE BLUFF, AR
^£ Lagrue Bayou
>•
Mississippi River
TEXARKANA.TX-AR
White River
Major Urban Area
Major Water Body
State Boundary
Interstate Freeway
60
0
60
120
180 Kilometers
Figure 1. Geographic overview of the Lower White River Region (LWRR) and White River Watershed (WRW) study areas.
-------
Study Site
Orientation Maps
Figure 2a. Orientation map of Lower White River Region (LWRR), Mississippi Alluvial
Valley Ecoregion (MAVE), and White River Watershed (WRW). Inset shows the states
that intersect the MAVE study area.
Figure 2b. Orientation map of Lower White River Region (LWRR), including the 13
counties that intersect the LWRR study area.
Figure 2c. County map of all study areas including 170 counties in Arkansas, Missouri,
Louisiana, Tennessee, Illinois, and Kentucky that intersect the Lower White River
Region (LWRR), Mississippi Alluvial Valley Ecoregion (MAVE), and the White River
Watershed (WRW).
-------
60 0 60 120 Kilometers
State Boundary
White River Watershed Study Area
Lower White River Region Study Area
Mississippi Alluvial Vafley Ecoregion Habitat Study Area
Figure 2a.
-------
Mississippi River
0
Kilometers
90 180
Figure 2b.
-------
Lower White River Region
White River Watershed HUC Boundary
Mississippi Alluvial Valley Ecoregion
County Boundary
State Boundary
60 0 60 120 Kilometers
Figure 2c.
-------
f ] HUC Boundary
Land Cover (NLCD)
HJ Open Water
| | Low Intensity Residential
^^| High Intensity Residential
|^| Commercial/lndustrial/Transportation
Bare SoilBare Rock/Sand/Clay
Quarries/Strip Mines/Gravel Pits
Transitional
Deciduous Forest
Evergreen Forest
Mixed Forest
Deciduous Shrubland
Orchards/Vineyards/Other
Grasslands/Herbaceous
Pasture/Hay
Row Crops
Small Grains
Urban/Recreational Grasses
Woody Wetlands
Emergent Herbaceous Wetlands
40
40
80 Kilometers
Figure 3. Land cover (National Land Cover Dataset; NLCD) in the White River Watershed, showing subwatershed
boundaries (per USGS 8-digit Hydrologic Unit Code).
-------
BEFORE 1600
OR OTHER ORIGINAL TYPES CURRENT (1992)
ND, PASTURE, OR URBAN
MISSOURI
.J, _ _ . , __
-•Tir —•* ' ' "• •
Figure 4. Wetland conversion to crop agriculture, pasture, or urban land cover (pre-1600 to
1992) in the Lower Mississippi River Basin (The Nature Conservancy, 1998).
-------
Figure 5. "Loading logs on a lumber company railroad in early 1900" in an Arkansas bottomland hardwood
swamp (Photo: Arkansas Historic Commission). Deforestation and draining of swamps allowed land to be
farmed or developed for other uses..
-------
Figure 6. Diesel generator pumping groundwater for the irrigation of row crops in Arkansas County, Arkansas (Photo:
Ricardo D. Lopez, 2001).
-------
Figure 7. The Beouf-Tensas Project, Arkansas-Louisiana (Photo: U.S. Army Corps of Engineers, date unknown), an
example of riparian vegetation destruction and wetland hydrologic alteration in the Mississippi Alluvial Valley Ecoregion.
-------
Figure 8. The McClellan-Kerr Navigation Channel on the Arkansas-Louisiana border (Photo: Ricardo D.
Lopez, 2001), an example of riparian vegetation destruction and wetland hydrologic alteration in the Lower White
River Region (LWRR).
-------
Lower White
River Region
Landsat Images
(November, 1999)
Figure 9a. Lower White River Region reference image with selected towns indicated
with blue arrows; Landsat ETM+ single near-infrared band; 30 meter spatial resolution;
November, 1999. Red rectangle indicates habitat vulnerability focus area in the South
Unit of the White River National Wildlife Refuge.
Figure 9b. Lower White River Region reference image with selected towns indicated
with blue arrows; Landsat ETM+ 3 band false-color infra-red composite; 30 meter spatial
resolution; November, 1999. Red tones indicate photosynthetic vegetation. Red
rectangle indicates habitat vulnerability focus area in the South Unit of the White River
National Wildlife Refuge.
Figure 9c. Lower White River Region reference image with selected towns indicated
with blue arrows; Landsat ETM+ 3 band false-color 'enhanced vegetation' infra-red
composite; 30 meter spatial resolution; November, 1999. Green tones indicate
photosynthetic vegetation. Red rectangle indicates habitat vulnerability focus area in the
South Unit of the White River National Wildlife Refuge.
-------
N
120 150
Grubbs, AR
Brinkley, AR
Clarendon, AR
Holly Grove, AR
Stuttgart, AR
DeWitt, AR
Figure 9a.
-------
N
0 3D 60 90 120 150
Kilometers
Grubbs, AR
McCrory, AR
^^^EKft-rV
JK. iirT
rMJt
L£F-'; -;;*
Des Arc, AR.
Brinkley, AR
Clarendon, AR
Holly Grove, AR
Stuttgart, AR
DeWitt, AR
I
K
Figure 9b.
-------
N
30 SO 90 120 150
Kilometers
McCrory, AR
• * i.*»?vf%B
.'••.jfc • -
v>^>'j.j£'»-
Grubbs, AR
W;
DesArc, AR.!
. :^
Brinkley, AR
Clarendon, AR
Holly Grove, AR
Stuttgart, AR
*n|;~-
\
H^
DeWitt, AR
il
Figure 9c.
-------
Figure 10. Land cover in the White River Watershed study area was measured adjacent
to surface water bodies and within each cumulative 30 m riparian zone. Single blue line
indicates edge of water. Each colored band represents a selected length of riparian zone
from the shoreline to a maximum of 300 m.
-------
Habitat Suitability Models
Habitat and Water Quality Vulnerability Models
Lower White River Basin Model
' X
Taxa quick
reference image
Habitat suitability
map name and
legend / Habitat suitability
map
Mississippi Alluvial Valley Ecoregion Model
Figure 11. How to interpret the habitat
suitability, habitat vulnerability, and water
quality vulnerability maps in this report.
itj-rt-
Grayscale
image
reference
guide
County
boundary
reference
guide
Habitat vulnerability
map name
and legend,
based on standard
deviations from the mean
vulnerability parameter
value
"8-digit HUC"
Watershed Model
White River
Watershed
water quality
vulnerability
models
Riparian Zone
assessment 90 m
maps at 30 m
interval
\
Landscape 180 m
metric name
and legend,
based on
quintiles
\
8-digit HUC
assessment
map
-------
M1
1 III
POP ROAD NWI AR-GAP MA
1 1 1 1
III 1
M15 M16 M12 M13 M2
III I
I I I
M17 M14 M3
I I I
M'
I
8 M4 IV
WV-LULC
5
I I
M6 M7
I I
,
I
M8 M
I
M"
M19
I
9 M10
|
;1
Figure 12, Hierarchical schematic of the mallard duck (M) winter habitat modeling process for
the Lower White River Region (LWRR). See Table 1a for information about data components,
Table 2a for habitat suitability model parameters, and Table 3a for habitat vulnerability para-
meters. M1 = Study site boundaries; POP = Population Block Group Data; ROAD = Wessex
Road Data; NWI = National Wetland Inventory Data; AR-GAP = Arkansas GAP Data; MAVA-
LULC = Mississippi Alluvial Valley of Arkansas Landuse/Landcover; M2 = 15 class mallard
duck winter habitat suitability under current conditions (all land, all sizes included); M3 = 10
class mallard duck winter habitat suitability under current conditions (wetland habitat only),
patches of all sizes; M4 = 10 class mallard duck winter habitat suitability under current
conditions (wetland habitat only), patches less than 2 ha; M5 = 10 class mallard duck winter
habitat suitability under current conditions (wetland habitat only), patches > 2 ha; M6 =
Wetland habitat patch total area, patches > 2 ha; M7 = Wetland habitat patch total perimeter
length, patches > 2 ha; M8 = Wetland habitat patch interior-to-edge ratio, patches > 2 ha;
M9 = Wetland habitat patch sinuosity index, patches > 2 ha; M10 = Wetland habitat patch
circularity index, patches > 2 ha; M11 = Wetland habitat unified patch index, patches > 2 ha;
M12 = Wetland habitat patch road density, patches > 2 ha; M13 = Wetland habitat patch total
road length, patches > 2 ha; M14 = Wetland habitat patch road index, patches > 2 ha; M15 =
Wetland habitat patch human population density in 1990, patches > 2 ha; M16 = Wetland
habitat patch human population density in 2011, patches > 2 ha; M17 = Wetland habitat patch
human population density change from 1990 to 2011, patches > 2 ha; M18 = Wetland habitat
patch unified human index, patches > 2 ha; M19 = Wetland habitat patch unified vulnerability
index, patches > 2 ha.
-------
Figure 13. An "emergent wetland" (per Cowardin et al., 1979) in the Lower White River Region (Photo: Ricardo
D. Lopez, 2001).
-------
Figure 14. A "scrub/shrub wetland" (per Cowardin et al., 1979), across a pool with duckweed (Lemna sp.) and
watermeal (l/Vo/ffiasp.) on the surface; Lower White River Region (Photo: Ricardo D. Lopez, 2001),
-------
*
Figure 15. A "forested wetland" {per Cowardin et al.. 1979), in the Lower White River Region (Photo: Ricardo D. Lopez, 2001}.
-------
Figure 16. (a) An "unconsolidated bottom wetland" (perCowardin et al., 1979), across an open field and (b) in an
"oxbow wetland", which is an isolated pond that is formed from an abandoned meander of a river (Photo: Ricardo
D.Lopez, 2001).
-------
Figure 17. An "aquatic bed" (per Cowardin et al., 1979) with submersed vegetation present in the foreground.
Emergent vegetation and forest present across an area of open water (Photo: Ricardo D. Lopez, 2002).
-------
Figure 18. A portion of the White River in the Lower White River Region study area (Photo: Ricardo D. Lopez, 2001).
Rivers provide areas for animals to feed, drink, and rest.
-------
Upland Palustrine Upland Palustrine
Upland
Palustrine
Upland
I a temporarily flooded d intermittently exposed
b seasonally flooded e permanently flooded
^^
high water
average water
low water
Figure 19. Features and examples of classes and hydrologic modifiers in cross-sectional view of palustrine
wetlands (from Coward in et al., 1979)
-------
Figure 20. Narrow sandy shores on the banks of the White River in the Lower
White River Region study area (Photo: Ricardo D. Lopez, 2001).
-------
Lower White River
Region Habitat
Suitability Models
-------
Figure 2la. Lower White River Region (LWRR) mallard duck winter habitat suitability
(All Land), All Sizes.
Figure 21b. Lower White River Region (LWRR) black bear habitat suitability (All
Land), All Sizes.
Figure 21c. Lower White River Region (LWRR) least tern habitat suitability reference
aid (All Land), All Sizes.
Figure 2Id. Lower White River Region (LWRR) least tern habitat image overlay (All
Land), All Sizes.
Figure 21e. Lower White River Region (LWRR) mallard duck winter habitat suitability
(Wetlands Only), All Sizes.
Figure 2If. Lower White River Region (LWRR) black bear habitat suitability
(Wetlands Only), All Sizes.
Figure 21g. Lower White River Region (LWRR) wetland plant habitat suitability
(Wetlands Only), All Sizes. Classes and hydrologic modifiers per Cowardin et al.
(1979).
Figure 21h. Lower White River Region (LWRR) mallard duck winter habitat suitability
(Wetlands Only), < 2 ha.
Figure 21i. Lower White River Region (LWRR) black bear habitat suitability (Wetlands
Only), < 2 ha.
Figure 21j. Lower White River Region (LWRR) wetland plant habitat suitability
(Wetlands Only), < 2 ha. Classes and hydrologic modifiers per Cowardin et al. (1979).
Figure 21k. Lower White River Region (LWRR) mallard duck winter habitat suitability
(Wetlands Only), > 2 ha.
Figure 211. Lower White River Region (LWRR) black bear habitat suitability (Wetlands
Only), > 2 ha.
Figure 21m. Lower White River Region (LWRR) wetland plant habitat suitability
(Wetlands Only), > 2 ha. Classes and hydrologic modifiers per Cowardin et al. (1979).
-------
Lower White River
Region Habitat
Suitability Models
(All Land, All Sizes)
-------
'Jki
H
..- -c
Figure 21a.
Mallard Winter Habitat Suitability (All Land) . All &zes
^^| Fluctuating hydrapenod wetland, trees.''shrubs. oak, noovercup oah
^^| Fluctuating hydrapehod wetland, trees-'E-hrubs.. oak, overcup oak
^B Fluctuating hyxlropedcxl wetland, trees/shrubs, no oafc
Fluctuating hydraperiod wetland, herbaceous plants.
Fluctuating hyxinopefiod wetland, no plants
^^| FiareJyflooded wetland , Lake, Impoundment, Rivsr, or Stream, trees/shrubs, oak, noovercup oak
^| Rarely ftooded wetlatxJ, Lake. Impoundment, River, or Stream, trees/shrubs, oah. overcup oah
^B Rarely Hooded wetland , Lake. Impoundment, River, or Stream, trees/shrubs, no oak
Rarely Hooded wetlanid . Lake. Impoundment, River, or Sfream, nefOaceous plarvis
Ranalyflooded wetland , Lake, Impoundment, River, or Stream, no plants
Agriculture (Com)
Agriculture (Rtee|i
Agriculture (Mb)
Agriculture (Soy)
BB Non-agricullural upland
-------
Figure 21 b.
Black tea- 1 labilal Su-iatiilsy • All Land. An Siies
IWotidy wufland Wiln 4 InM apeCJM ptMant prior 10 SSStTa
Woodywoaswd w-lh 3irw spacwt prwwt pnorlo s95Crs
Woody wofland win 3 Into spec** prttaanc prior 10 195CTH
Woody wafland win '. Inw spaces prwtm prior 1o 19SCT*
B Wittily upland ™ui. 4 tie* siMtoua prftMCil pnor 10 1ft5ffs
Woody upland with 3 trea epecm pre&eni prior b 1 GWs
Wixidy uptana witn 2 [raft afi«CHM pras«nl (ma lo 155tf a
W«ody upland •Mtu 1 Eroft BpAoM prosKni pnor cci 1 Kffa
Woody wefland -win -4 irm spaces prewnt afl*r 196CTB
WnMdy wauamd Mlh 1 ITM spuciM praurn afior tgscrt
Woody weflond w-ih ? ln» SPKJM pr^wit afl** I»WB
Wdody wodand wllh 1 Inw aptews pi*s*nc aluM 1950's
Woody upland wtn 4 tree &PWBS prsswii altor iftWs
Weody upland «4irv 3 craa fipeom jx-aMni artw 19Sffi
Woody upland with 2 (WB toecm preoart aftw 11KWs
Woody upland «4Elr1 [Wt aposes presenl alCut ISSffa
Nwi-vmudy >Mt»c4anil wift 4 puni ipociSH pf«aort. non-bf ewiec n tno 19SO'a
Non-woody wotiand wiOn 3 p&m spodei prownc, non-fweelefl n UH iSSO'fl
Non-weedy wrtland win 2 pUnl apndera pfflfldM, non-hxealoO n the 1950'S
Non-woody watiamt wir 1 planl spodM presaM. non-fcresletf n ttw 1950'H
N&n-wOddy wMlBnif w«i[h D pBnl SpBaeS pMSMC, nOn-f^fiftlfid n tin t950'B
Non-woody upland with 4 obm speoos pr*M^i no^Tor«s(ed -" the 1-&$f>i
Nfin-wnKidy upland with a planl siMCto prtscnl. ittn-rorMUd 4n flw 1 KPa
Non-woody upland with 7 oiani ep«Mi pn>&om_ non-fofflstod m the iiHta
ih t piiiril spetes. prustril in>p,-[cnt«ed in tM 1%ff«
in 0 plaril speuirt pnJaWil iion-IOrealad in DM 155ffa
Non-wOOdy nMlamJ wflfi 3 ptanl speuus prMOi't. 'LO^tlLfJ in ln» -.950*»
Non-woody wadantt wtn 2 psni spaa« preewt, formbd in the *9 Ws
Non-wo^dy wdund win 1 pfeni ipedaa pfutru, terMttd in ifnr 195Cf»
Non-woody -wtianfl *iji o piani spsoes prwert. 'oreAd In iht iSWs U«
Non-woody upland wltf^ 4 pbnt«|»£iDa^«HrYi fomtud *i Bw 19S(n f^.s..
Nsf i-v,uady uplUld wiflt 2 ptanl spetjes pntBunl fihealBS in 0*6 IKfft (>.<-..
Non-woody upland with i jjianl ep
-------
Figure 21 c.
Suitable For Tern Nesting
Unsuitable For Tern Nesting
-------
Figure 21d.
Suitable For Tern Nesting
-------
Lower White
River Region
Wetland Habitat
Suitability Models
(All Sizes)
-------
N.
Figure 21e.
Mallard Winter Habitat Suitability {Wetlands Onty), AH Sizes
• Fluctuating hydroperiod wetland, trees/shrubs, oak, no overcup oak
Fluctuating hydropenod wetland, ireea'shrobs. oak, o^ercup oak
F luctua'ing hydrDpennd wetland, Irees/shrubs, no oak
FJuetuatir*g hydroperiod wetland, herbaceous plants
FluctuBlrng hydroperiod wetland, no plants
| Rarely flooded weltand , Lake,
j Rarely flooded wetland , Lake,
j Rarely Hooded welland , Lake,
Rarely flooded wetland , Lake,
Rarely flooded welland , Lake,
ipoundment. River, or Stream, irees-'sftrubs, oak, no overcup oak
ipoundmenl. River, or Stream, trees/shrubs, oak, overcuD oak
mpoundmenl, River, or Stream, trees/shrubs, no oak
ipoundment. River, or Stream, hertaeeous plants
iDDundmenl, River, or Stream, no plants
-------
VUh.4
id with 4 frw species rswUpnor to 1960's
WixsJywetlaKlwith.atr'eeapeciea reaant prior to 195Q's
WoMly wHUattf wtth 2 tree species «wit prior to 195a'a
Woody wetland with 4 tnw specie* pmwnl after 195CTs
iwcotfy wfiUand with a free apecies
Woody v..9iuir• e, d
I Non-wooiIy wetland m 2 plantapeciea ftesaa. fweatea n tha 1950's (>e . diatufted)
|Non^T70dywqtiaridwnnOpiani«pK«({nB9itfDfwfeHlfiti« I950'«(ic. d
-------
Figure 21g.
Wertland Plan! Matxlai Suability (WeSlands Qn\y). A! 5t/es
HI |L] LacufHrino, [2] Littoral, [AS] Aquaiic Bod. [3] Rooted Vase Jar, [F] SemiperniBnemly Rooded
H [L] Lscuslnne. [2] Littoral. [AS] Aquatic Bed. [3] Rooted Vaaedar, [H] Permanently Floodeo
|L] Lacu&nrKJ. [2] L moral, [UB| UnconsrintaSBd Bottom, [H] PerrnanenUy Flooded
B[P] (Mlualnna. [AB] Aquatic Bed. [3] Rooted Vascular, |H| PermonerMly Flooded
I'-'J Palualnne [EM] Emergent, [1] Perstsjant, [A[ Temporanly Flooded
=== ]P] PaKislrine. (EM] EmergenL [t] Petstflent. JC] Seasonally Flocded
^BJ |P] Paluslnne-, [EM] Emeroant. |1] P&'sisiant, [F] Se^iipermanefilly Flooded
I IP] I
| [Pj Palustnn
| JP] Patuslnr
| |P] Paluslrin
j JP] Pafustnn
![P]Pahis1rin
|P] Pafufiln
;A. Temporanly Flooded
[C] Seasonally Flooded
JF] SemipennanenUy Flooded
i. [Cj Soasonaty Flooded
;. [Ft SempennanQnttv Flooded
'ealed. [1| Broad-Leaved DeodLH
f. (FO| vesled. [1 [ Brmd-Leaved Dectdut
i:[FOJ oreated.[lj Broad-Leaved Deodui
L [FQJ orested. P) N
t (FQj oresled. [2j Needte-Leaved C
9, EFO[ oreated, (ty Deciduous, [A] Tecnporanly Ftooded
£. [FO! oresred. [6} Dedduous.. 1C] Seasonally Roodcd
is [SS] Scrub-Shrub. [>] Broad-Leaved Deoduous. [A] Temporanty Flooded
| [PJ Paluslnne (SSj Scrub-Shrub. [1J Broad-Leaved Deoduoua, [Cf SeasonaJfy Flooded
| pj Paluslriro. [SS] Scrub-Shrub, [1] B^nad-Leaved Deciduous, [F] SenxjemianenUy Flooded
| JP] Paluslnne (UBi Unconsobdatad Bcflom
![P] Paluslrine. (US| UnoonsolxtelBd Snore. (C] Seasonally Rooded
|R[ Riurjnno. |2| Loww ParonnigJ. (UB! UnconsoUtglwJ Bottom, fH[ PermanenDy Flooded
jftj Rivenne. [5| Lower PafBfi/isal, [US' Uneonsobdaied Sfxve. [AJ Temooranly Flooded
| |RJ Rivonne. 12| Lower Pw^nnigl. [US! UnooncoMa^ed Shore, EC] Seasonally Flooded
-------
Lower White
River Region
Wetland Habitat
Suitability Models
(< 2 ha)
-------
.4 .i
Figure 21h.
Mallard Winter Habitat Suitability {Wetlands Onty), less than or equal to 2 hectares
• Fluctuating hydroperiod wetland, trees/shrubs, oak. no overcup oak
Fluctuating hydroperiod wetland, ireea'sftrubs. oak. o^ercup oak
F luctua'ing hydrDperind wetland, Irees/shrubs. no oak
FJuetuatir>g hydroperiod wetland, herbaceous plants
FluctuBltng hydroperiod wetland, no plants
| Rarefy flooded weltand , Lake,
j Rarely flooded wetland , Lake,
j Rarely Hooded welland , Lake,
Rarely flooded wetland , Lake,
Rarely flooded welland , Lake,
ipoundment. River, or Stream, irees-'shrubs, oak, no overcup oak
ipGUndmenl, River, or Stream, trees/shrubs, oak, overcuD oak
mpoundmanl, River, or StrBain, trees/shrubs, no oak
ipoundmenl. River, or Stream, nertiaceous plants
ipoundmenl, River, or Stream, no plants
-------
VUh.4
-t (•
Figure 21 i.
* Bear Wfltlartd Kabila! TLu itnhiky - 'iVcllands iess :hjn cir equ
!W«By wfUind *iwn frm) species pnnwnt juior ID iffiB's
Wuottjr wtahvid wim 3 tree species present pricr U ! 950's
Woxly vrfriland with ? 1rm affldea present jwlw EO t950's
Wootfy *«1iand win t tw spedn present fwicr ED iSSO's
WqoUy welicmd wm J mw SJMCJUS prawnuflor l
-------
Figure 21j.
Wetland Plant Habilat Suitability (Wetlands Only), less than or equal to 2 hertates
Ml I1-] Lacustrine. I?] Littoral, |A9] Aquatic Bed. |3] Hoolod Vascular. [F] Semipennarwntty Flcoded
H [L] LaeusEnne. [2] Littoral. [A3] Aquatic Bed. [3] Rooted Vascular, [H] Permarwnlly flooded
PJ LaoMlrirw, |2] LUtoral. |UB| Unconsolidated Bodwn, |H[ Pgrmanantty Fl™dBd
B|Pj F>alualnne [AQ] Aquale Bed, |3] Rootad Vascular. |H| Permanently Flooded
p] Pakialrina, [EM] EjiMfgBnt, |1] PersstenL [A| TempofarJy Flooded
|P] Paluslnrio. [EM| Emergent f1] Pensfetenl. [C] Seasonally Floodad
S|P] PaJuatnne-. [Efv^ EntefgenU jl] Peraisterrl,. [F] SflimpenranenUy Flooded
{P^l Paluslnns, [FOJ Foraated, [1| Eroad-Laawad DaodLfflUEk [A| Temporarily Flooded
IP] PaKislrfna. [FOf FWaslBd. [1| Enoad-LeavOd Deciduous. [C] Saasonally Flooded
^B JPJ Paluslnne. [FQ| hcresled [1| Broad-Leaved Deoduous, [F] $ernipermarwntiy Flooded
E| |P| PiiiuElnnu (KM Frjrrjslcd [2| Neodl9-Leav«) Oociduous. |C| Stsasonallv Flooded
H lpl Paluslnnrj [FO| Forested. [2] NeiodIff-Leaved Deoduws. |F| SemipflnnanonUv Floodsd
B| |pj f'aluslnne [F0| Forested, [B| Deciduoua [A] Temporanty Flooded
B[P] Palustrirw. [FOJ Fonrelad. [6| Dacidwous. [C] Seasonally Flooded
|P] f>aluslnne [SS] Scrub-Shrub. 11] Bfoad-Leaved D&ciduoos. [A] Tamporanly Fknded
BW P] F>aluslnne. [SS] Scrub-Shrub. |1] Broad-Leaved Deciduous. |C| Seasonally Flooded
•V lpl Paluslnne. [SS] Scrub-Shrub, [1] Broad-Leaved Deciduous, |F| Somipennanentlv Flooded
HI |P] Paluslnna. [UB| UnconsolidatKJ Hotloni
B|P] PaJuslnne. [US| Unccnsolidated Shofw. [C] Seasonally Flooded
|R| Rivghr>rj, [2| Low«r Perennial. [UB| Unconsolidatod Sotlom, [H| Pennanently Flooded
|Hj Rivenrre, [2] Lo^er Perennial [US| Unconsolidafea Snore-. [A| Temporarily Flooded
•• |R| Rivr.nno. 12| Lowr Pgrorimgl. [US] UrconsoTidafed Shore. [C] SeaMmaJty Floodod
-------
Lower White
River Region
Wetland Habitat
Suitability Models
(> 2 ha)
-------
Figure 21k.
Mallard Winter Habitat Suitability {Wetlands Only), > 2 hectares
• Fluctuating hydroperiod wetland, trees/shrubs, oak. no overcup oak
Fluctuating hydropenod wetland, trees.'sftnjbs. oak. overcup oak
F luctua'ing hrydropen'od wetland, Irees/shrubs. no oak
FJuetuatir*g hydroperiod wetland, herbaceous plants
FluctuBlrng hydroperiod wetland, no plants
| Rarely flooded weltand , Lake.
j Rarely flooded wetland . Lake.
j Rarely Hooded welland . Lake.
Rarely flooded wetland , Lake.
Rarely flooded welland , Lake.
ipoundment. River, or Stream, irees-'shrubs, oak, no overcup oak
ipoundcnenl. River, or Stream, trees/shrubs-, oak, overcua oak
mpoundmanl, River, or StrBain, trees/shrubs, no oak
npoundmenl. River, or Stream, neffiaceous plants
ipoundriienl, River, or Stream, no plants
-------
VUh.4
tbMy - WkMarrta > Z heaanes
. Wcody walBfld with 1 v«e a*eaea pnawnt prior to 1B5Q1*
Wootfy wHUand with 2toe s^naaaa present prior to 1B5D1
WoKiy*eilafldwth4fri»5j»aes rewntaflor i*5Crs
!Woo{tyhVEi1l&>ndwiin3veflapQda9 rewnt aflar 195CTa
Wowty wellei^ with ? tea sfMcfn rewnt sfler I^Ws
tee n»d« resent afler ISSffs
h 4 plant aptOM j>«a«vL
-------
Figure 21m.
Wetland Plgni Hatxlal Suability (Wcllands Only), > 2 hectors
BlLl Lacus^rinn. [Z] Littoral, [AS] Aqu-aiic Bod. 13] Rooted Vascular, [F] Semiperrnanertly Fl
|LJ Lacuslnne. [2] Lil:ora!. LAB] Aquauc Bad. [3] Rooted Vascular. [H] Pe?manentfy flooded
[I] Lacustrine. [2] Littoral. [UB| Unccnaciidaleo: Bottom, [H) Parniaremiy Flooded
B|F>] (Miijsinne [Afi] Aqcaiic Bad, [3] Rooted Vascular, [H| Perma^efHly Flooded
jf-'J f'aiuslnne (€M\ EmargenU [1] Persisjant, :.A| Tanporanly F(ooded
|P] Paluslrino. (EM] EmergenL [1] PorEiSBnt. [C] Seasonally FtoodwJ
S[P] Paluslnna, [EM] EmafgenL, |1] Persislant, JF] Semiparmanefilly Flooded
[P] Paluslrino. [FO] Foreslod. [1| Broad-Learod Deciduous. {Aj Temporarily FoowxS
JP] Paluslnno. [FO] ForesLwt. [1| Broad-Lsarod Deciduous. (C] Seasonally F^xxWd
^B I^J f'aluslnna £FO] Forested, [1| Broad-Laaued Deciduous, [f] SemipennanenUy Flooded
B(P] PiUuslnng [FO] Fomsted. [Z| Nefldte-teavod Deciduous, [Cj Seasonal? Flooded
P] Pahratrirw, [FO] Forested. [2\ Ne«d'c-LB3vod Dectfucxis. [Ft Sen^pennangntrv Flooded
BB IP] Patualnne., [FO] Forested, [fij Deciduous, £Aj Tempofanly Flooded
B[P] PBlustine. [FO] FoFBSted. [61 Deciduous. fCJ Seasonally Fboded
I'-'J Palualnne. [SS] Scruti-Shrub, [1] Srcad-Leavad Deoduous, [A] Tarraoraey Flooded
H jf'j f'aluslnne (S£j Scrub-Shrub, |1 J Braad-Leaved Oecxhxxis, [CJ Seaaonaly Flooded
|P] Psluslrme. [SS] Scrub-Shrub. [1] B^ced-Leavgd Deciduous, [F] Senwerrnanenpy Flooded
i P]
[P]
|R| Rivrjn
[R[ Riveri
IR1
. .
(UB| Unoortsobd^lad Bottom
. [USJ Unconsolxlal^d Store. [Cl Seasonally Rooded
. \2\ Lxnwr Perennial. [UB] Unconsoddalod BdtOcn, [H[ Pormanenlly FbOdod
, [^ Lower PfifKifuaT, [US] Uneonsotetalad Shore, (A{ Temporanty Flooded
, 1 2| Lower Perennial, [US] Uncon&obdaJod S^oro. [C] Seasonally Flooded
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Patch Size and
Patch Shape)
-------
Figure 22a. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total area.
Figure 22b. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total area.
Figure 22c. Lower White River Region (LWRR) wetland plant habitat vulnerability (>
2 ha) in terms of habitat patch total area.
Figure 22d. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total perimeter length.
Figure 22e. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total perimeter length.
Figure 22f. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch total perimeter length.
Figure 22g. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch interior-to-edge ratio.
Figure 22h. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch interior-to-edge ratio.
Figure 22i. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch interior-to-edge ratio.
Figure 22j. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch circularity index.
Figure 22k. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch circularity index.
Figure 221. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch circularity index.
Figure 22m. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch sinuosity index.
Figure 22n. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch sinuosity index.
Figure 22o. Lower White River Region (LWRR) wetland plant habitat vulnerability (>
2 ha) in terms of habitat patch sinuosity index.
Figure 22p. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch unified patch index.
Figure 22q. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch unified patch index.
Figure 22r. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch unified patch index.
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Total Area)
-------
Figure 22a.
Mallard Winter Habitat - Total Area, patches > 2 hectares
IB -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
m > 3 Std. Dev. (Least Vulnerability)
-------
VUh.4
•»•;-..
^ <•
•^-*» j« i .
>;t v -.«r»
•:.jN
;-^rx
"f«
*--l
Black Bear Wetland Habitat • Total Area, patches > 2 hectares
^B -1 - 0 Sid. Dev. (Greatest Vulnerability)
Mean
0 - 1 SU. Dev.
1 - 2 Sid Dev.
B 2 - 3 Sid Dev.
> 3 sid. Dav. (Least Vulnerability)
-------
Wetland Plant Habitat - Total Area, patches > 2 hectares
M -1 - 0 Sid. Dev. (Greatest Vulnerability)
Mean
0 -1 Sid Dev.
1 - 2 SW. Dev.
2 - 3 Sid. Dev
^H > 3 Std. Dev. (Least Vulnerability)
- t
bfo
S
H
JT_
* If
••N ' . *
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Total Perimeter)
-------
Figure 22d.
Mallard Winter Habitat - Total Perimeter Length, patches > 2 hectares
^H -1-0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std, Dev.
1 - 2 Std. Dev,
2 - 3 Std. Dev,
•• > 3 Std. Dev. (Least Vulnerability)
-------
tnmUf Length, pattfo! > J i«c
0 - 1 Sid. Dav.
1-2SU. Day.
.2-3 Std. Day.
>3SM nev i.L
-------
Figure 22f.
Wetland Plant Habitat • Total Perimeter, patches > 2 hectares
H -1 -OSW, Dev (GreatestVulnerability)
Mean
D - 1 Std. Dev.
1-2SK). Dev,
2 - 3 Std. Oev.
HI > 3 Std Dev. (Least Vulnerability)
.
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Interior-to-Edge
Ratio)
-------
Mallard Winter Habitat - Interior to Edge Ratio, patches > 2 hectares
^B -2 - -1 Std. Dev. (Greatest Vulnerability)
•B -1 - 0 Std. Dev.
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^H > 3 Std. Dev. (Least Vulnerability)
-------
Black Bear Wsfland Habital • Interior B Edss Ftalo, patehea * 2 heclares
^| -2 • -1 Strt, D«V. Ifircn^at V^lrcraailiiy I
^ -t-OSW. Dm.
Mean
0 • I Std- Beu
1 -2 Std. Da».
2-3SUJ Oev
^| > 3 Std DeV CLenoI VuhlBrBbtuy:i
-------
v.i
4
Wetland Piani Habitat -Inietior to Edfle Raiio, patedes > 2 heoares
•• -2 - -1 Sid. Dev. (Greatest Vulneradiily)
^B -1-0 Sid Oev.
Mean
0 - 1 Std Dev.
1 - 2 Std. Dev.
2-3 Std. Dev.
^| > 3 Sid, Dev (Least Vulnerability)
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Circularity Index)
-------
.*;•
Mallard Winter Habitat - Circularity Index, patches > 2 hectares
IB -1-0 Std, Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^| > 3 Std. Dev. (Least Vulnerability)
T«
;
-------
VUh.4
Slack Gear Wellana Habilal - Circularity Index, palcnes > 2 Iwdaras
•I • i - r, siii n«v (Creates! Vulnerability)
Mean
0-tStd. Dev-
1-SSM.Dm.
S2-3SM. Dev
> 3 su. Dev. (
-------
. «*••
Wetland Plant Habitat - Circularity Index, patches > 2 hectares
^H -1 - 0 Sid Dev (Greatest Vulnerability)
Mean
0 -1 SW Dev.
1 - 2 Stt. Dev
2-3 Sid De»
^H > 3 Sid. Dev, (Least Vulnerability)
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Sinuosity Index)
-------
Mallard Winter Habitat - Sinuosity Index, patches > 2 hectares
^| -2 - -1 Std. Dev. (Greatest Vulnerability)
- i - 0 Std. Dev.
Mean
0 -1 Std. Dev.
1 -2 Std. Dev.
2-3 Std. Dev.
^| > 3 Std. Dev. (Least Vulnerability)
-------
VUh.4
Black Ereir Wetland HaOila! - Sinuosity Index, patches > 2 he Hares
H -1-0 Sid. Dev. (Greatest Vulnerability i
Mean
0-IStd. Dev.
1-2Std. Dev.
B2-3SW. Dev.
> 3 SW. Dev. (Least Vulnerability)
-------
/
Wetland Plant Habitat - Sinuosity index, patches > 2 Hectares
^B -1 - 0 Sid Dev (Greatest Vulnerability)
Mean
0 -1 SM Dev.
1 - 2 Stt. Dev
2-3 Sid De»
^H > 3 Sid. Dev, (Least Vulnerability)
?f-zm
1 *" L^-" * -•'"".»""
• •
^fe
•ft.*
;«-'
7'
j*.
-•^*
**5
<-tat-
*.v-^
*-'-
'*:
:/X
4.
<
-•
^"'?
I*.*,
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Unified Patch Index)
-------
v -: -
. *".-.-
•f-sJv,?
•
Figure 22p.
Mallard Winter Habitat - Unified Patch Index, patches > 2 hectares
^B -1 - 0 Std. Dev, (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^| > 3 Std. Dev. (Least Vulnerability)
-------
VUh.4
*f:
r?**
> „•.«
••'+*•»-'
«;'-:-*--i
^v
^-
BIB* Bssr Wetona Habllat • United Pach Index, palches * 2 Hectares
^B -1 ' OStd &ev iflrralc^l y'i:-Ifrl.lbi rv
Mean
0-1 Std. Cm.
1 - S Sid. Ca^.
S 2 - 3 Sid. Dev-
> 3 Std. Oav. iL«a.tvu»*n»tjl
-------
JH»
Wetland Plant Habitat - Unified Patch Index, patches > 2 hectares
^H -1 - 0 SW Dev (Greatest Vulnerability)
Mean
0 -1 SW Dev.
1 - 2 Stt. Dev
2-3 Sid De»
^H > 3 Sid. Dev, (Least Vulnerability)
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Patch Human-
Induced Disturbance)
-------
Figure 23a. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total road length.
Figure 23b. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total road length.
Figure 23c. Lower White River Region (LWRR) wetland plant habitat vulnerability (>
2 ha) in terms of habitat patch total road length.
Figure 23d. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total road density.
Figure 23e. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total road density.
Figure 23f. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch total road density.
Figure 23g. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch road index.
Figure 23h. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch road index.
Figure 23i. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch road index.
Figure 23j. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch human population density in 1990.
Figure 23k. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch human population density in 1990.
Figure 231. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch human population density in 1990.
Figure 23m. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch estimated human population density in
2011.
Figure 23n. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch estimated human population density in
2011.
Figure 23o. Lower White River Region (LWRR) wetland plant habitat vulnerability (>
2 ha) in terms of habitat patch estimated human population density in 2011.
Figure 23p. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch human population density change from
1990 to 2011.
Figure 23q. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch human population density change from
1990 to 2011.
Figure 23r. Lower White River Region (LWRR) wetland plant habitat vulnerability (> 2
ha) in terms of habitat patch human population density change from 1990 to 2011.
-------
Figure 23s. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch unified human index.
Figure 23t. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch unified human index.
Figure 23u. Lower White River Region (LWRR) wetland plant habitat vulnerability (>
2 ha) in terms of habitat patch unified human index.
Figure 23v. Lower White River Region (LWRR) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch unified vulnerability index.
Figure 23w. Lower White River Region (LWRR) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch unified vulnerability index.
Figure 23x. Lower White River Region (LWRR) wetland plant habitat vulnerability (>
2 ha) in terms of habitat patch unified vulnerability index.
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Total Road Length)
-------
Mallard Winter Habitat - Total Road Length, patches > 2 hectares
H -1 - 0 StcJ. Dev. (Least Vulnerability)
Mean
0 -1 Std, Dev.
1 - 2 Std, Dev.
2 - 3 Std. Dev.
^B > 3 Std- Dev- (Greatest Vulnerability)
-------
VUh.4
Black Bear Wstond Habitat - Total Road Length, palettes > 2 hectares
M -1 - 0 Sta. Dm (Leas! Vulnerability)
Mean
0 -1 SM. Dav.
1-ZSMDn.
z-SStd-Dev.
^>3SW Dew. (Greatest Vulnerability)
-------
WeUand Plant Habitat - Road Length, psioies > 2 hectares
HI -1 - 0 stu Dev (Least Vu neramny)
Mean
0-1 Sid. Dev
1 - 2 Sid Dev
2- 3 Sid Dev
HI > 3 Std. Dav. (Greatest Vutn*rab*h1v)
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Total Road Density)
-------
Mallard Winter Habitat - Road Density, patches > 2 hectares
iB -1 - 0 std- Dev- (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^H > 3 Std, Dev. (Greatest Vulnerability)
-------
M
Black Bear Wetland Habitat - Road Density, patches •• 2 hectares
H -i - 0 Sta. Di-v (Least Vulnerability)
Mean
0-1Std Dev
1-2 Sid Oev.
2 - 3 Sid. Dev
^ > 3 sm Dev (Greatest Vulnerability)
V
'%'
-------
Wetland Plant Habitat - Road Density, patches > 2 hectares
H| •' - 0 Sto Dev (Leasl Vulnerability)
Mean
0-1 Sid Dev
1 -2Sld-D«v
2-3Sld.D«u.
^| > 3 Sid Dev (Greatest Vulnerability)
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Road Index)
-------
Mallard Winter Habitat - Road Index, patches > 2 hectares
^| -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 -2 Std. Dev.
2 - 3 Std. Dev.
H| * 3 Std. Dev. (Greatest Vulnerability)
-------
VUh.4
BBtk BaarWelland Habitat - Road Index, pattties > 2 hectares
Hi -1-0 Std. Qav. (Least Vulnerability)
Mean
O-lStd.Dev.
1-2SUJ. Dev.
2-3 Sid. Dev.
H = 3 Std. Dev. (Greatest Vulnerability)
-------
Wetland Plant Habital - Road Index, patches > 2 hectares
^H -1 - o sir: FIWV (Least Vulnerability)
Mean
0 -1 S(d, Oev
1 - 2 Sid. Dev.
2-3SW Oev
^| > 3 SIS Do» (Greatest Vulnerability)
•*;
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Human Population
Density in 1990)
-------
Mgllart Winter Habitat - Human Population Density in 1990, patches > 2 hectares
^H -1 - 0 Sid. Dev. (Least Vulnerability)
Mean
0 - 1 Std. Dev.
1-2Std. Dev.
2-3Std. Dev.
^B > 3 Std. Dev. (Greatest Vulnerability)
VT--
v
' ' ->
j I •
-------
Slack SMT Wfltwnd HatNtiU - Human Pojunban Cwnsti' In 1990. j»ltf*s > 2 «
H -1-OSW.Cw (LfiJStVLinsrabiityl
Maun
0-1SW Dev
1 - 2 SW Dtrv.
2-3SM. Dw.
-------
WeHand Plan! HabiLat - Human Population Density in '990. catenas > 2
H -! - 0 Sid. Dftv LLflasl VUri*rgb.iilv!
Mean
0- 1 S*d CW
1 -2SM.Dev.
2 - 3 SKI D*v.
^ ' 3 Sid. D»V (Grwtfis! Vulnerability^
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Estimated Human
Population Density
in 2011)
-------
Mgllart Winter Habitat - Human Population Density in 2011, patches > 2 hectares
^H -1 - 0 Sid. Dev. (Least Vulnerability)
Mean
0 - 1 Std. Dev.
1-2Std. Dev.
2-3Std. Dev.
^B > 3 Std. Dev. (Greatest Vulnerability)
-------
fc But* Wetland HgWal - Human PopuJalen 0*»ity m »T 1, petehM > 2
-i-OSld.Oflw [Leaal Vi^nerabHyi
MHO
0-ISd.Ow,
l-^SHDttt
I-3SB Oev
-------
-
Oana Plan HatnLat - Human Population Damity in 20i t, patches > 2
' • - D &0- CX|tf (l-BKll V\llri«fg Lullty'l
Mean
0-1 SM Duv.
1 -2SH.Dev.
2 - 3 SW. D*v.
' 3 SW. Dwv (Gra»tB5! Vulnerab«)i^J
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Estimated Human
Population Density
Change from
1990 to 2011)
-------
Figure 23p.
Mallard Winter Habttal • Human Popijlalkn Density Change from 1990lo2011,palches> 2 hectares
HH < -3 Std- Dev- t^Eiaal Vulnerability}
^ -3 - -2 Sid. Dav.
-2--1 Std.Dev,
iB i O^g FiO'.
Mean
0 -1 Std. D&v.
1 -1 Std. Dev.
•1 ^ 3 Sid Dev.
•J > 3 Std. Dev. <;Greases! Vulnerabilnv)
-------
:nOmrifr Qwi9»r«m IBflOtoKil1, pvtEhM - ?r«
.
Mun
0 I 511 D.IV
1-?SM Dov
J-3SM D*v
•
-------
•-/", '-f
l Plant HaWal - Humgn Popyladon Density Cfiange from 1^0 to 201V
p ' ..'.
ft
0 - 1 5W
1 -25B
?- 3SW
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Unified Human
Index)
-------
Mallard Winter Habitat - Unified Human Index, patches > 2 hectares
HB < -3 Std. Dev. (Least Vulnerability)
HH -3 - -2 Std. Dev.
-2--1 Std, Dev.
H -1 - 0 Std. Dev.
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
IB 2 - 3 Std. Dev.
•• > 3 Std. Dev. (Greatest Vulnerability)
-------
VUh.4
Black Bear Wetland Hatala: - trailed H uman Indsx. patches > 2 hectares
•I < .3 sw. Dev- (Least Vulnerability)
• -3-2SM-DBV,
-2--iaa DSV.
^ • Lisa Dev,
Mean
0-1SM. Dev,
1-2SU, Dev,
B2 - 3 S«. Dn.
>3SM. Dm. (Oreatesl Vulnerabiliiy)
'•£\
•X
1
X.
-*>•'
. s.
c£
£S
K,^.
--,-,_;>;-!
\
we 5
;>-:•*
-------
Wflland Plant Hatrtal - Uniliad Human Index, pachas - 2 hectares
H -- 3 gld Dov (Loil Vunorab'^;,
0 - 1 5W
1 -25B
?- 3SW B
-------
Lower White
River Region
Wetland Habitat
Vulnerability Models
(Unified Vulnerability
Index)
-------
v
Mallard Winter Habitat - Unified Vulnerability Index, patches > 2 hectares
JBI f -3 std- Dev- (Greatest Vulnerability)
B -3 - -2 Std. Dev,
-2 - -1 Std. Dev,
-1 - 0 Std. Dev.
Mean
^H 0 • 1 Std. Dev.
1 - 2 Std. Dev.
H 2 - 3 Std. Dev.
^H > 3 Std. Dev. (Least Vulnerability)
-------
VUh.4
sat* Bwr wenano nalital - un« VtiiwMnlily MB. poteJws > Iiwctam
^| < -3 Std DBV. vG'ealeil Vulne'ab.llyl
•I •> < S!H n.!,
-t-OSId Dm
Mean
1-ZSW D»
2-3SU Dev.
^ > 3 Std. DBV ,
if,,-
-------
I -t -03W •>.
Mew
0 - 1 5W D«v.
1 - 2 5M Dev.
? - 3 SB D»v
- .,
r
s
ly Index, patches =- 2 hectares
-------
Mississippi Alluvial
Valley Ecoregion
Habitat Suitability
Models
-------
Figure 24a. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
suitability (All Land), All Sizes.
Figure 24b. Mississippi Alluvial Valley Ecoregion (MAVE) black bear habitat
suitability (All Land), All Sizes.
Figure 24c. Mississippi Alluvial Valley Ecoregion (MAVE) least tern habitat suitability
reference aid (All Land), All Sizes.
Figure 24d. Mississippi Alluvial Valley Ecoregion (MAVE) least tern habitat image
overlay (All Land), All Sizes.
Figure 24e. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
suitability (Wetlands and Open Water Only), All Sizes.
Figure 24f. Mississippi Alluvial Valley Ecoregion (MAVE) black bear habitat
suitability (Wetlands and Open Water Only), All Sizes.
Figure 24g. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
suitability (Wetlands and Open Water Only), All Sizes.
Figure 24h. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
suitability (Wetlands and Open Water Only), < 2 ha.
Figure 24i. Mississippi Alluvial Valley Ecoregion (MAVE) black bear habitat
suitability (Wetlands and Open Water Only), < 2 ha.
Figure 24j. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
suitability (Wetlands and Open Water Only), < 2 ha.
Figure 24k. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
suitability (Wetlands and Open Water Only), > 2 ha.
Figure 241. Mississippi Alluvial Valley Ecoregion (MAVE) black bear habitat
suitability (Wetlands and Open Water Only), > 2 ha.
Figure 24m. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
suitability (Wetlands and Open Water Only), > 2 ha.
-------
Mississippi Alluvial
Valley Ecoregion
Habitat Suitability
Models
(All Land, All Sizes)
-------
Figure 24a.
Mallard Winter Habitat Suitability (All Land), All Sizes
B| Wetland, non-open water, trees or shrubs
Wetland, non-open water, herbaceous plants
Non-wetland, agriculture
^B Open water
^^ Non-wetland, no agriculture
-------
Figure 24b.
Btacfc Sear Wetland Habitat Suitability (All Land) , All Sizes
^H Wetland, Trees* <25Qm from inee
Wetland, Non-agricultural Herbaceous, <25Qm from tree
Wetland., Nan-agricultural Herbaceous. >25Qm from tree
Upland, Trees, <250m from iree
Upland, Non-agricultural Herbaceous, <£50m from tree
Upland. Non^agricultural Herbaceous. >25Dm from tree
^H Upland, Agriculture. Urban, or other IVon-vegetaled, <25Qm from iree
^H Upland, Agriculture, Urban, or other r4on-vegetated, >250m from iree
^| OpenWaler
-------
Figure 24c.
Suitable For Tern Nesting
Unsuitable For Tern Nesting
-------
Figure 246.
Suitable For Tern Nesting
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Suitability Models
(All Sizes)
-------
Figure 24e.
Mallard Winter Habitat Suitability (Wetlands and Open Water Only) . All Sizes
^B Wetland, non-open water, trees or shrubs
Wetland, non-open water, herbaceous plants
^B Open water
-------
Figure 24f.
Black Bear Wetland Habital Suitability (Wetlands and Open Water Only), All Sizes
HI Wetland, Trees, <250rn from Iree
Wetland, Non-agricultural Herbaceous. <25Qm from tree
Wetland, Non-agricultural Herbaceous. >250m from tree
HI Open Water
-------
Figure 24g.
Wetland Plant Habitat Suitability (Wetland Only), All Sizes
B Woody Wetlands
Emergent Herbaceous Wetlands
H Open Water
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Suitability Models
(< 2 ha)
-------
Figure 24h.
Mallard Winter Habitat Suitability (Wetlands and
Open Vfeler Only), tess than or equal to 2 hectares
|^B Wetland, non-open water trees or shrubs
Wetland, non-open water, herbaceous plants
^H O:;-e" water
-------
Figure 24L
Black Bear Wetland Habitat Suitability (Wetlands and
Open, Water Only), less Irian or equal to 2 hectares
^H Wetland, Trees, <25Qm from tree
I Wetland. Non-agricultural Herbaceous, *25Dm from tree
Wetland. Non-agriculturai Herbaceous. >250m from tree
•I Open Water
-------
Figure 24j.
Wetland Plant Habitat Suitability (Wetland Only)
less than or equal to 2 hectares
^H Woody Wetlands
Emergent Herbaceous Wetlands
^| Open Water
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Suitability Models
(> 2 ha)
-------
Figure 24k.
Mallard Winter Habilat Suitability (Wellands and Open Water Only), > 2 hectares
^m Wetland, non-open water, trees or shrubs
Wetland, non-open water, herbaceous plants
j^B Open water
-------
Figure 241.
ear Wetland Habitat Suitability (Wetlands and Open Water Only), -> 2 hectares
Wetland, Trees, <2§Qrn from tree
" Wetland. Mon^agricullural Herbaceous. <.25Qm from tree
Wetland. Non-agricultural Herbaceous, >2SOm from tree
Open Water
-------
Figure 24m.
Wetland Plant Habitat Suitability (Wetland Only), > 2 hectares
^H Woody Wetlands
Emergent Herbaceous Wetlands
^B Open Water
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Patch Size and
Patch Shape)
-------
Figure 25a. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total area.
Figure 25b. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total area.
Figure 25c. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch total area.
Figure 25d. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total perimeter length.
Figure 25e. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total perimeter length.
Figure 25 f. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch total perimeter length.
Figure 25g. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch interior-to-edge ratio.
Figure 25h. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch interior-to-edge ratio.
Figure 25i. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch interior-to-edge ratio.
Figure 25j. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch circularity index.
Figure 25k. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch circularity index.
Figure 251. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch circularity index.
Figure 25m. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter
habitat vulnerability (> 2 ha) in terms of habitat patch sinuosity index.
Figure 25n. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch sinuosity index.
Figure 25o. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch sinuosity index.
Figure 25p. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch unified patch index.
Figure 25q. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch unified patch index.
Figure 25r. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch unified patch index.
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Total Area)
-------
Figure 25a.
Mallard Winter Habitat - Total Area, patches > 2 hectares
|H -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
|B 2 - 3 Std. Dev.
HI =* 3 Std. Dev- {Least Vulnerability)
-------
V? Figure 25b.
Black Bear Wetland Habitat, Total Area, patches > 2 hectares
^B -1 - 0 Std Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
j^B > 3 Std. Dev. (Least Vulnerability)
-------
Figure 25c.
Wetland Plant Habitat, Total Area, patches > 2 hectares
^H -1-0 Std. Dev. (Greatest Vulnerability)
Mean
0-1 Std. Dev.
1 -2 Std. Dev.
2 - 3 Std. Dev.
HI =• 3 std- Dev. (Least Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Total Perimeter)
-------
Figure 25d.
Mallard Winter Habitat - Total Perimeter Length, patches > 2 hectares
Hi -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 - 1 Std, Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^| > 3 Std. Dev. (Least Vulnerability)
-------
Figure 25e.
Black Bear Wetland Habitat, Total Perimeter Length, patches > 2 hectares
^H -1-0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev,
1 - 2 Std. Dev.
2 - 3 ad. Dev.
^| > 3 Std. Dev. (Least Vulnerability)
-------
Figure 25f.
Wetland Plant Habitat, Total Perimeter Length, patches > 2 hectares
• -1 - 0 Std. Dev, (Greatest Vulnerability)
Mean
0 -1 Std. Dev,
1 - 2 Std. Dev.
2 - 3 Std. Dev.
B > 3 Std. Dev. (Least Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Interior-to-Edge
Ratio)
-------
Figure 25g.
Mallard Winter Habitat - Interior to Edge Ratio, patches > 2 hectares
BB -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0-1 Std. Dev.
1 - 2 Std, Dev,
2-3Std. Dev.
BH > 3 Std. Dev. (Least Vulnerability)
-------
Figure 25h.
Black Bear Wetland Habitat, Interior to Edge Ratio, patches > 2 hectares
^H -1-0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^B > 3 Std. Dev. (Least Vulnerability)
-------
Figure 25i.
Wetland Plant Habitat, Interior to Edge Ratio, patches > 2 hectares
^B -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 - 1 Std. Dev.
1 - 2 Std. Dev.
B2 - 3 Std. Dev.
> 3 Std. Dev. {Least Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Circularity Index)
-------
Figure 25j.
Mallard Winter Habitat - Circularity Index, patches > 2 hectares
IB -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
| > 3 Std. Dev. (Least Vulnerability)
-------
Figure 25k.
Black Bear Winter Habitat, Circularity index, patches > 2 hectares
HI -1 - 0 Std, Dev. (Greatest Vulnerability)
Mean
0 -1 Std, Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^H > 3 Std. Dev. (Least Vulnerability)
-------
Figure 251.
Wetland Plant Habitat, Circularity Index, patches > 2 hectares
^| -1 - 0 Std, Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 -2 Std. Dev.
2-3 Std. Dev.
^f > 3 Std. Dev. (Least Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Sinuosity Index)
-------
Figure 25m.
Mallard Winter Habitat - Sinuosity Index, patches > 2 hectares
m -1 - ° Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
i(H > 3 sw- Dev- (Least Vulnerability)
-------
Figure 25n.
Biack Bear Wetland Habitat, Sinuosity Index, patches > 2 hectares
HI -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
d. Dev. (Least Vulnerability)
-------
Figure 25o.
Wetland Plant Habitat, Sinuosity index, patches > 2 hectares
~f -1 -OStd Dev.(Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
t 2-3SW. Dev.
H > 3 Std. Dev (Least Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Unified Patch Index)
-------
Figure 25p.
Mallard Winter Habitat - Unified Patch Index, patches > 2 hectares
^| -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std, Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^H > 3 Std Dev. (Least Vulnerability)
-------
3n$i Figure 25q.
Black Bear Wetland Habitat, Unified Patch Index, patches > 2 hectares
^H -1 - 0 Std. Dev. (Greatest Vulnerability)
Mean
0 -1 Std Dev.
1 - 2 Std. Dev.
2 - 3 Std, Dev.
^| > 3 Std. Dev. (Least Vulnerability)
-------
Figure 25r.
Wetland Plant Habitat, Unified Patch Index, patches > 2 hectares
^| -1-0 Std, Dev. (Greatest Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2-3 Std. Dev.
> 3 Std. Dev. (Least Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Patch Human-
Induced Disturbance)
-------
Figure 26a. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total road length.
Figure 26b. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total road length.
Figure 26c. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch total road length.
Figure 26d. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch total road density.
Figure 26e. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch total road density.
Figure 26f. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch total road density.
Figure 26g. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch road index.
Figure 26h. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch road index.
Figure 26i. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch road index.
Figure 26j. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch human population density in 1990.
Figure 26k. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch human population density in 1990.
Figure 261. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch human population density in 1990.
Figure 26m. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter
habitat vulnerability (> 2 ha) in terms of habitat patch estimated human population
density in 2011.
Figure 26n. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch estimated human population density in
2011.
Figure 26o. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch estimated human population density in
2011.
Figure 26p. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch human population density change from
1990 to 2011.
Figure 26q. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch human population density change from
1990 to 2011.
-------
Figure 26r. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch human population density change from
1990 to 2011.
Figure 26s. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch unified human index.
Figure 26t. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch unified human index.
Figure 26u. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch unified human index.
Figure 26v. Mississippi Alluvial Valley Ecoregion (MAVE) mallard duck winter habitat
vulnerability (> 2 ha) in terms of habitat patch unified vulnerability index.
Figure 26w. Mississippi Alluvial Valley Ecoregion (MAVE) black bear wetland habitat
vulnerability (> 2 ha) in terms of habitat patch unified vulnerability index.
Figure 26x. Mississippi Alluvial Valley Ecoregion (MAVE) wetland plant habitat
vulnerability (> 2 ha) in terms of habitat patch unified vulnerability index.
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Total Road Length)
-------
Figure 26a.
Mallard Winter Habitat - Total Road Length, patches > 2 hectares
^H -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
|H * 3 std- Dev- (Greatest Vulnerability)
-------
Figure 26b.
Black Bear Wetland Habitat, Total Road Length, patches > 2 hectares
^| -1 - o Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^B > 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 26c.
Wetland Plant Habitat, Total Road Length, patches > 2 hectares
^B -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
IB " 3 std- Dev- (Greatest Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Total Road Density)
-------
Figure 26d.
Mallard Winter Habitat - Road Density, patches > 2 hectares
^B -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
H > 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 26e,
Black Bear Wetland Habitat, Total Road Density, patches > 2 hectares
f-1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 - 1 Std. Dev,
1 - 2 Std. Dev.
2 - 3 Std. Dev,
• > 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 26f.
Wetlanci Plant Habitat, Total Road Density, patches > 2 hectares
!•( -1 - 0 Sin. Dev, (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
^B > 3 Std, Dev. {Greatest Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Road Index)
-------
Figure 26g.
Mallard Winter Habitat - Road Index, patches > 2 hectares
|H -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
HI > 3 Std. Dev. (Greatest Vulnerability)
-------
Vf Figure 26h.
Black Bear Wetland Habitat, Road Index, patches > 2 hectares
^H -1 - 0 Std. Dev, (Least Vulnerability)
Mean
0-1 Std. Dev,
1 - 2 Std. Dev.
2 - 3 Std. Dev.
H > 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 26i.
L
Wetland Plant Habitat, Road Index, patches > 2 hectares
^B -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 -2 Std. Dev.
2-3 Std. Dev.
^B > 3 Std. Dev. (Greatest Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Human Population
Density in 1990)
-------
Figure 26j.
Mallard Winter Habitat - Human Population Density in 1990, patches > 2 hectares
j^B -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std Dev.
2 - 3 Std. Dev,
^B » 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 26k.
Black Bear Wetland Habitat. Human Population Density in 1990, patches > 2 hectares
JH -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 - 2 Std. Dev.
2 - 3 Std. Dev.
•i > 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 261.
Wetland Plant Habitat, Human Population Density in 1990, patcnes > 2 hectares
•I -1 - 0 Std. Dsv. (Least Vulnerability)
Mean
0-1 Std, Dev.
1 - 2 Std, Dev.
2 - 3 Std. Dev.
^| > 3 Std. Dev. (Greatest Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Estimated Human
Population Density
in 2011)
-------
Figure 26m.
Mallard Winter Habitat - Human Population Density in 2011, patches > 2 hedares
^f -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1 • 2 Std, Dev.
2 - 3 Std. Dev.
^| > 3 Std, Dev. (Greatest Vulnerability)
-------
I Figure 26n.
Black Bear Wetland Habitat, Human Population in 2011, patches > 2 hectares
^B -1 - 0 Std Dev. (Least Vulnerability)
Mean
0-1 Std. Dev.
1 -2 Std. Dev.
2 - 3 Std. Dev.
^m > 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 26o.
Welland Plant Habitat, Human Population in 2011, patches > 2 hectares
H -1 - 0 Std. Dev. (Least Vulnerability)
Mean
0 -1 Std. Dev.
1-2 Std. Dev.
2 - 3 Std. Dev.
BB > 3 Std. Dev, (Greatest Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Estimated Human
Population Density
Change from
1990 to 2011)
-------
Figure 26p.
Maffard Winter Hattlat - Human Popukalnyi Defistty Change from 1990 In 2011. patches > 2 hectares
j^B < -S SLtl Dev. (Least Vulnefabiltty)
j^B -3 - -2 Sid. Dev.
-2--1 Sid. Dw.
IB-1 -OStd.Dev.
Mean
0 -1 Sid. Om.
I - 2 Sid Dev.
IB 2 - 3 Sid. Dev.
HI '' ^ SU. Dev. (Greatest Vuliwrability)
-------
Figure 26q.
Black BearWellanct Habita!, Human Population Density Change from 1990 to 2011, patches
H " -3 Sid Dcv (Least Vulnerability)
^-3--2Sld.Oev.
•2--1 S(d.Dcv.
^B -1-0 std. Civ
Mean
0-1SW. Dav.
1 - 2 Std, Dm.
BS - 3 Sid Ow.
-• 3 Std. Dav. .1 j-«:lcu VuliHrabillty)
-------
Figure 26r.
Wetland Plant Habttal, Human Populalldn Density Change from 1990 to 2011. patches > 2 hectare
^| <-3Sld. Dev. (LeastVulnerability)
H ,3 - -2 Std Dev.
-2 - -1 Sw. Dev.
^ -1 -OSId
0- ISW.Dev.
l-2SM,Dev.
2 - 3 Std, Dev
> 3 Std. De». (Greatest Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Unified Human
Index)
-------
Figure 26s.
Mallard Winter Habitat - Unified Human Index, patches > 2 hectares
Bl * -3 std Dev (Least Vulnerability)
H -3--2 Std. Dev.
-2--1 Std. Dev.
H -1 - 0 Sid. Dev.
Mean
0 -1 Std. Dev.
1 - 2 Std Dev.
^B 2 - 3 Std. Dev.
^| > 3 Std. Dev. (Greatest Vulnerability)
-------
Figure 26t.
Black Bear Wetland Habital, Unified Human Index, patches > 2 hectares
^H < -3 Std. Dev. (Least Vulnerability)
BJ-a --S SW Beu
-2--1 Std. Dew.
^B -1 • OSId. Dm.
Moan
0-1 Sid. Dev
1-2 ad. Deu.
^H 2 - 3 Sid Dev.
^| > 3 Std. Oev. (Greatest Vulnerability)
-------
Figure 26u.
Wetland Plant Habitat. Unified Human Index, patches > 2 hectares
^H < -3 Std Dev. (Least Vulnerability)
M -3 • -2 Std. Dev.
-2 - -1 Std. Dev.
^B -1 - 0 Std. Dev.
Mean
0-1 Std. Dev.
1 - 2 Std. Dev.
- 3 Std. Dev.
> 3 Std. Dev. (Greatest Vulnerability)
-------
Mississippi Alluvial
Valley Ecoregion
Wetland Habitat
Vulnerability Models
(Unified Vulnerability
Index)
-------
Figure 26v.
llard Winter Habitat - Unified Vulnerability Index, patches > 2 hectares
< -3 Std. Dev. (Greatest Vulnerability)
-3 - -2 Std Dev
-2--1 Std. Dev.
-1 - 0 Sid. Dev
Mean
0 -1 Std. Dev.
1 - 2 Std, Dev.
I ^ - 3 Std. Dev.
B > 3 Sid. Dev (Least Vulnerability)
-------
Figure 26w.
Black Bear Wetland Habitat. Unified Vulnerability Index, patches =• 2 hectare
BB t -3 Std. Dev. (Greatest Vulnerability)
H-3--2 Std Dev.
-2--1 Std. Oev.
• 1 • 0 Std Dev.
BB 0-1 Sid. Dev.
1 - 2 Std Dev
BB 2 • 3 Sid. Dav.
BH s 3 std> Oev> C-easl Vulnerability)
X
-------
Figure 26x.
Wetland Plant Habitat, Unif ed Vulnerability Index, patches > 2 hectares
j^B < -3 Std, Dev (Greatest Vulnerability)
-2--1 Std. Dev,
-1 - 0 Std. Dev.
Mean
1 - 2 Std. Dev.
^B > 3 Std. Dev. (Least Vulnerability)
-------
White River
Watershed Water
Quality Vulnerability
Models and Figures
-------
11010007
11010006
11010002
\
8020402
8020401
8020204
- 8020201
11010009
11010012
8020302
8020203
8020205
8020304
8020100
8020303
Figure 27. Twenty-five 8-digit hydrologic unit code (HUC) subwatersheds in the White River
Watershed. Yellow triangle indicates a National Water Quality Assessment (NAWQA) Program
field sampling location, sampled during the 1990s (N = 35).
-------
HUC Boundary
Stream
40 0 40 80 120 160 200 240 280 320 Kilometers
Figure 28. Location of surface water in the White River Watershed. Surface water locational data was used to
calculate the percent of streams in proximity to roads.
-------
j HUG Boundary
/Road
S
40 0 40 80 120 160 200 240 280 320 Kilometers
Figure 29. Road network in the White River Watershed. Data used to calculate the percent of streams in
proximity to roads.
-------
Percent Streams Within 30 meters of a Road
| 0.01 - 0.03
0.03 - 0.04
| 0.04 - 0.04
0.04 - 0.05
0.05-0.07
Figure 30. White River Watershed percent of streams within 30 meters of a road.
-------
0 m
30m
60m
120m
150m
210m
240m
270m
300m
HUC
Percent Total Agriculture
JH 8 - 30
| | 30 - 46
| J46-61
f 1 61 - 76
^m 76 - 92
Figure 31. Proposed water quality vulnerability metrics, based on percent
agriculture (i.e., NLCD codes 81, 82, 83, and 85 in Table 4) adjacent to
shorelines, within cumulative thirty-meter riparian zones, and within entire
HUCs.
-------
0 m
30m
60m
90m
120m
150m
180 m
210m
240m
270m
300m
HUC
Percent Crop Agriculture
18-35
35-52
52-69
69-86
Figure 32. Proposed water quality vulnerability metrics, based on percent
crop agriculture (i.e., NLCD codes 82, 83, and 85 in Table 4) adjacent to
shorelines, within cumulative thirty-meter riparian zones, and within entire
HUCs.
-------
0 m
30m
60m
120m
210m
270m
300m
HUC
Percent Pasture Agriculture
0- 12
12-23
23-33
33-44
44-55
Figure 33. Proposed water quality vulnerability metrics,
based on percent pasture agriculture (i.e., NLCD code 81
in Table 4) adjacent to shorelines, within cumulative
thirty-meter riparian zones, and within entire HUCs.
-------
s
40 0 40 80 120 160 200 240 280 320 Kilometers
Figure 34. Hillshaded digital elevation model (DEM) of the White River Watershed, depicting areas from
lower elevation (darker) to higher elevation (lighter). This DEM was used to calculate percent of total
agrculture on slopes greater than three percent.
-------
Percent Total Agriculture on Slopes Greater than 3%
•MS-26
I 26 - 29
29 - 59
59-80
80-92
Figure 35. White River Watershed percent total agriculture on slopes greater
than three percent.
-------
11010007
8020202
11010002
11010C06
8020402
B020401
8020204
8020201
8020304
8020100
6020303
Figure 36. Spatial analysis of twenty-five 8-digit hydrologic
unit code (HUC) subwatersheds in the White River
Watershed. There is a relatively greater risk of surface
water quality impairment in 2 HUCs (outlined in red above
and at right) as a result of the presence of agriculture within
the cumulative 120 meter riparian zones.
270m
Percent Total Agriculture
H8-30
• 30-46
y46-61
61-76
• 76-92
300m
60m
HUC
-------
Largest Forest Patch Proportion of HUC
Hi 3.0051 - 14.8873
y 14.8873-50 1804
SD. 1804-72 1136
72.1136-843746
• 84 3746-8S 7723
Forest Patch. Mean Area (m J
Ml 35Q6.7S3 • 120B3.S43
~~l 12083.543 -27280.116
~* 27280.116- 170900.063
• 170800 063 • 231298.667
• 231296.667 • 520171.605
Forest Patch. Largest (m3)
B128790D-33041700
3304170D • 235755900
^] 2357559DD - 14097195DD
I 14097195DD-27115515DO
• 2711551500-5112D873DO
Fore6.1 Patch Density (No
| 1.6152-3.1934
J3.1S34-3.B9ei
J3.99B9-5.1S17
—I 5.1517-6.6975
I 6.8S7S • B.D026
ForeKt Patch Number (No.)
^13098-11165
J 113S9- 12B3S
\ 13291 - 17DDO
| 17326-24417
I 2E421 -62425
Foresl, Proportion of HUC
JiB 2.4982 - 6 2251
J 6^251 - 15.9727
15.9727-67.5539
| 67.5539 - 72.6923
• 72.6923-84.0197
Figure 37. Proposed water quality vulnerability metrics, based on (a) largest forest patch proportion of HUC, (b) mean area of forest patch,
(c) largest forest patch area, (d) forest patch density, (e) forest patch number, and (f) percent of HUC that is forest (by area). Forest cover
based on NLCD codes 41, 42, and 43 (Table 4).
-------
0 m
30m
60m
90m
120m
150m
180 m
210m
240m
270m
300m
HUC
Percent Forest
2-19
19-35
35-51
51 -68
68-84
Figure 38. Proposed water quality vulnerability metrics, based on percent
forest (i.e., NLCD codes 41, 42, and 43 in Table 4) adjacent to shorelines,
within cumulative thirty-meter riparian zones, and within entire HUCs.
-------
0 m
210m
240m
270m
300m
Percent Wetland
0- 13
13-27
27-40
40-53
53-67
Figure 39. Proposed water quality vulnerability metrics, based on percent
wetland (i.e., NLCD codes 91 and 92 in Table 4) adjacent to shorelines,
within cumulative thirty-meter riparian zones, and within entire HUCs.
-------
0 m
30m
60m
90m
120m
150m
180 m
210m
240m
270m
300m
HUC
Percent Natural Land Cover
7-22
22-38
38-54
54-69
69-85
Figure 40, Proposed water quality vulnerability metrics, based
on percent natural land cover (i.e., NLCD codes 31, 41, 42, 43,
51, 71, 91, and 92 in Table 4) adjacent to shorelines, within
cumulative thirty-meter riparian zones, and within entire HUCs.
-------
8020202
11010CD2
11010C36
8020204
8020201
11010009
11010012
8020302
8020203
802020S
80203M
8020100
8020303
Figure 41. Spatial analysis of twenty-five 8-digit hydrologic
unit code (HUC) subwatersheds in the White River
Watershed. There is a relatively greater risk of surface
water quality impairment in 6 HUCs (outlined above and
at right) as a result of forest loss in riparian zones of various
widths.
0 m
;
90 m
180m
270m
Percent Forest
•12-19
L19-35
35-51
51 -68
• 68-84
30 m
120m
210m
300m
60m
150m
HUC
-------
Lower White River
Region Federal
Wildlife Refuge
Analyses
-------
B
C
A
— ' JL,
ear L^ 71
69 /* 7
70 72
Figure 42. Seventy-two federal refuge zones in the Lower White River Region were compared using mallard
duck winter habitat vulnerability models. Federal refuge zones are all within a larger Lower White River Region
study area. Overview image (upper left) is separated into three areas to show the location and identification
number of each federal refuge zone (enlarged in images, A-C). Because of its larger area, federal refuge zone
72 is shown both in image quadrant B (northern zones), and in image quadrant C (southern zones).
-------
Mallard Winter Habitat - Unified Vulnerability Index, patches > 2 ha
^H < -3 Std. Dev. (Greatest Vulnerability)
| -3 - -2 Std. Dev.
-2 - -1 SW. Dev.
-1 - 0 Std. Dev.
Mean
HH 0 • 1 Std. Dev.
1 - 2 Std. Dev.
^| 2 • 3 Std. Dev.
H| =• 3 Std. Dev. (Least Vulnerability)
Figure 43. The unified vulnerability index was applied to a hypothetical
landscape where change in the riparian wetland hydrology has occurred.
Red rectangle indicates the portion of the South Unit in the White River
National Wildlife Refuge where landscape change is hypothesized.
-------
| (,: -3 Sid Dp*.} .
I (-3 10 •<> SM. Osw.)
(-2IO-1SUJ. Dev.)
(•1 loOSM, Dgv)
MHO
I (3 la 1 Sill Oev |
(t -•.!.•. 5: i 7;'.v;
I (S lo 3 Sid Ocv,}
I (> 3 Std. Own.) =
A
Kilometers
202468
N
Figure 44. (A) Mallard duck winter habitat vulnerability under current hydrologic conditions in the South Unit of the White River
National Wildlife Refuge; (B) Loss of mallard duck winter habitat under hypothetical decrease in flood stage and duration on the
White River; (C) enlarged view of predicted net loss of mallard duck (winter) habitat under hypothetical decrease in flood stage
and duration on the White River.
------- |