Partners in Flight

Bird Conservation Plan

for the

Mississippi Alluvial Valley

(Physiographic Area # 05)

Version 1.0

September 1999

Written by: Daniel Twedt, David Pashley, Chuck Hunter, Allen Mueller, Cindy Brown, and Bob Ford

Address comments to:

Bob Ford, c/o U.S. Fish and Wildlife Service, University of Memphis, South Campus, Bldg 8, Memphis, TN 38152

901-327-1752; fax 901-327-8001; bford@tnc.org

ACKNOWLEDGMENTS

Partners in Flight Bird Conservation Plans are living documents that are periodically updated and improved as knowledge of bird management and conservation in the physiographic area is improved. Thus the first public draft of this Bird Conservation Plan is Version 1.0. This Plan will remain a work in progress, we encourage those who wish to contribute to do so.

The following individuals participated in drafting or revision of this conservation plan: S. Ray Aycock, Charles K. Baxter, Cindy Brown, Robert J. Cooper, Lisa Creasman, Tom Foti, Robert P. Ford, Paul B. Hamel, Bob Hatcher, Doug Helmers, W. C. (Chuck) Hunter, Brad Jacobs, Sunni Lawless, Charles R. Loesch, Curt McMurl, Seth E. Moth, Allen J. Mueller, Gary Myers, David N. Pashley, Lance Peacock, Terry Rich, Karen Rowe, Janet Ruth, Jon Schneider, Mike Staten, Mark Swan, Daniel J. Twedt, William B. Uihlein, Bill Vermillion, Mark S. Woodrey, and Scott C. Yaich.

§ TABLE OF CONTENTS

• EXECUTIVE SUMMARY
• ACKNOWLEDGMENTS
• SECTION 1: THE PLANNING UNIT

o Background

o Conservation issues

o Conservation opportunities

• SECTION 2: AVIFAUNAL ANALYSIS

• Species prioritization
• Conservation area size considerations
• Minimum viable population and source populations

• SECTION 3: HABITATS AND OBJECTIVES

o Forested wetlands, bottomland hardwood forests

• Status and importance
• Priority species, species suites, and habitat requirements
• Habitat and population objectives
• Implementation recommendations and opportunities
• Evaluation of assumption

o Forest Openings, Edges, Early-Successional Shrub-Scrub

• Status and importance
• Priority species, species suites, and habitat requirements
• Habitat and population objectives
• Implementation recommendations and opportunities
• Evaluation of assumptions

o Upland Oak-Hickory Hardwoods

• Status and importance
• Priority bird species, species suites, and habitat requirements
• Habitat and population objectives
• Implementation recommendations and opportunities.
• Evaluation of assumptions.

o Grasslands/Savannas, Pastures, and Associated Wetlands.

• Status and importance.
• Priority species, species suites, and habitat requirements.
• Habitat and population objectives.
• Implementation recommendations and opportunities.
• Evaluation of assumptions.

o Lacustrine (open-water) Wetlands, Palustrine (emergent) Wetlands, and Mudflats.

• Status and importance.
• Priority species, species suites, and habitat requirements.
• Habitat and population objectives.
• Implementation recommendations and opportunities.
• Evaluation of assumptions.
• Urban, Suburban backyards, Rural Woodlots.
• SECTION 4: IMPLEMENTATION RECOMMENDATIONS AND SUMMARY.
• Summary of habitats, bird species suites, and goal statements.
• LITERATURE CITED

• Table 1. Native habitats and current land use of physiographic provinces within the Mississippi Alluvial Valley.
• Table 2. Hydrologic units within the Mississippi Alluvial Valley.
• Table 3. Priority bird species.
• Table 4. Hypothesized forest area (hectares) required to support viable populations of 500 breeding birds within the Mississippi Alluvial Valley.
• Table 5. Area of bottomland hardwood forest types.
• Table 6. Habitat objectives for predominately mature, forested wetlands.
• Table 7. Area in hectares of forested habitat within the Mississippi Alluvial Valley.
• Table 8. Area of shrub/scrub and grassland habitats (in hectares) within the Mississippi Alluvial Valley.
• Table 9. Area of upland forest (in hectares) on ridges within the Mississippi Alluvial Valley.
• Table 10. Twenty-five years (1966-1991) of Breeding Bird Survey population trends.
• Table 11. Shorebird species, average mass, and hypothesized abundance during southward migration through the Mississippi Alluvial Valley.
• Table 12. Proposed distribution of managed foraging habitat for shorebirds.
• Figure 1. Mississippi Alluvial Valley physiographic province.
• Figure 2. Bottomland hardwood forest habitat patch objectives.
• Figure 3. Proposed balance sheet to be used to track progress toward shorebird foraging-habitat objectives
• Appendix 1. Tree species
• Appendix 2. Breeding bird species priorities, percent of population (PP), area's importance (AI), population trend (PT), and typical habitat
• Appendix 3. Methodology used to establish foraging habitat objectives for shorebirds
• Appendix 4. Qualitative Characterization of Bird Conservation Areas in the Mississippi Alluvial Valley
• Appendix 5. Quantitative Characterization of Bird Conservation Areas (BCA)in the Mississippi Alluvial Valley

SECTION 1: THE PLANNING UNIT

Background:

The Mississippi Alluvial Valley (Fig. 1) is an 11 million ha (24 million acres), relatively flat, weakly dissected alluvial plain, comprised of natural levees, basins and flats, point bar formations, terraces, tributary floodplains, and depressional wetlands. Differences in topography and hydrology result in 14 physiographic provinces spanning 7 states (Table 1; Keys et al. 1995). Elevation ranges from 0 to 200 m (0-660 feet) with local relief generally <30 m but reaching up to 100 m along ridges and bluffs bordering the mainstem Mississippi River. Because elevation differences are slight, hydrologic regimes can dramatically influence vegetation. In addition to 21 internal hydrologic units (Table 2), the Mississippi Alluvial Valley receives drainage from many eastern and central U.S. watersheds, including the Ohio, Arkansas, and Red Rivers. Average annual precipitation is 114 to 165 cm (45-65 in).

Potential natural vegetation for most of the physiographic area is southern floodplain forest with oak-hickory forest on higher ground (e.g., Crowley's Ridge, loess bluffs) and isolated native prairies (e.g., Grand Prairie). Floodplain forests are primarily oak-gum-cypress cover type with co-dominant species being overcup, willow, Nuttall, water, swamp chestnut, and cherrybark oaks, as well as sweetgum, water tupelo, water hickory, willow, cottonwood, sycamore, hackberry, sugarberry, red maple, boxelder, baldcypress, and green ash (scientific names of trees listed in Appendix 1). Oak-hickory forests include as co-dominants post, southern red, black, and white oaks, and shellbark, shagbark, and mockernut hickories. Remnant prairies have bluestem (Andropogon spp.) and switchgrass (Panicum virgatum) as the dominant grasses under natural conditions. Natural vegetation has been cleared from 80% of this physiographic area primarily for conversion to agriculture. Pasture and haylands are common on higher ground and along levees. Cotton, soybean, and rice are the most widespread crops but winter wheat, corn, sorghum, and sugar cane can be locally abundant. Although cleared of natural vegetation, flooded agricultural fields can provide important wildlife habitat.

Flood regimes that have historically dictated vegetative communities within the Mississippi Alluvial Valley have been altered by an extensive system of levees, dikes, and dams. High water events have been reduced in many areas, whereas the rate and extent of flooding has been increased in other areas. The altered hydrology of the Mississippi Alluvial Valley has in turn influenced the composition and structure of forested wetlands as well as the amount of open water-shoreline habitat. Local perturbations include ice-storms, tornadoes, hurricanes, beaver ponds, and fire. The northern portion of the Mississippi Alluvial Valley overlays the New Madrid fault line with the potential for earthquakes.

Table 1. Native habitats and current land use of physiographic provinces within the Mississippi Alluvial Valley.
State	Physiographic Province	Native Habitats	Land Use
Arkansas	Southern Mississippi River Alluvial Plain	cottonwood - willow, oak - sweetgum, tupelo - cypress	cropland, forestry
	Crowley's Ridge	post - blackjack oak, southern red oak - white oak, beech - maple	pasture
	Arkansas Grand Prairie	willow oak - overcup oak, bluestem - switchgrass	agriculture
	Arkansas Alluvial Plain	overcup oak - sweetgum, green ash - American elm-hackberry - sugarberry	agriculture
	Macon Ridge	southern red oak - white oak - post oak - hickory [pignut, mockernut, sand]	agriculture, forestry
	Bastrop Ridge	southern red oak - white oak - post oak - hickory [pignut, mockernut, sand], shortleaf pine - oak [white, southern red, post, black]	agriculture, forestry
	North Mississippi River Alluvial Valley	river birch - sycamore, overcup oak - sweetgum	agriculture
	White and Black Rivers Alluvial Plain	oak - sweetgum, overcup oak - water hickory, tupelo - cypress	cropland
	St. Francis River Alluvial Plain	overcup oak - sweetgum, green ash - American elm - hackberry - sugarberry	agriculture
Illinois	North Mississippi River Alluvial Valley	river birch - sycamore, overcup oak - sweetgum	agriculture
Kentucky	North Mississippi River Alluvial Valley	river birch - sycamore, overcup oak - sweetgum	agriculture
Louisiana	Southern Mississippi River Alluvial Plain	cottonwood - willow, oak - sweetgum, tupelo - cypress	agriculture, forestry
	Baton Rouge Terrace	loblolly pine - oak [cherrybark, swamp chestnut, Shumard], shortleaf pine - oak [white, southern red, post, black]	agriculture
	Atchafalaya Alluvial Plain	water tupelo - bald cypress, overcup oak - sweetgum	agriculture
	Macon Ridge	southern red oak - white oak- post oak - hickory [pignut, mockernut, sand]	agriculture, forestry
	Red River Alluvial Plain	oak [swamp chestnut, cherrybark, Shumard] - sweetgum, green ash - American elm - hackberry - sugarberry	agriculture
	Bastrop Ridge	southern red oak - white oak - post oak - hickory [pignut, mockernut, sand], shortleaf pine - oak [white, southern red, post, black]	agriculture, forestry
	Opelousa Ridge	loblolly pine - oak [cherrybark, swamp chestnut, Shumard], oak - sweetgum	agriculture, forestry
	Teche Terrace	oak [swamp chestnut, cherrybark, Shumard] - sweetgum, green ash - American elm - hackberry - sugarberry	agriculture
Mississippi	Southern Mississippi River Alluvial Plain	cottonwood - willow, oak - sweetgum, tupelo - cypress	agriculture
	North Mississippi River Alluvial Valley	river birch - sycamore, overcup oak - sweetgum	agriculture
Missouri	Crowley's Ridge	post - blackjack oak, southern red oak - white oak, beech - maple	agriculture, forestry
	North Mississippi River Alluvial Valley	river birch - sycamore, overcup oak - sweetgum	agriculture
	White and Black Rivers Alluvial Plain	oak - sweetgum, overcup oak - water hickory, tupelo - cypress	agriculture
	St. Francis River Alluvial Plain	overcup oak - sweetgum, green ash - American elm - hackberry - sugarberry	agriculture
Tennessee	North Mississippi River Alluvial Valley	river birch - sycamore, overcup oak - sweetgum	agriculture

Conservation issues:

The Mississippi Alluvial Valley is among the most heavily modified physiographic areas in the southeastern U.S. but still supports the largest forested floodplain in North America. In addition to providing forested habitat for breeding birds, the area serves as a major waterfowl wintering area, supplying food and cover from both forested and agricultural habitats. Resident and migrant long-legged wading birds as well as transient shorebirds exploit natural floodwater, flooded farm fields, and aquaculture ponds.

Although habitats for waterbirds have not historically been a focus for Partners in Flight, we address habitat needs of nongame waterbirds because bird conservation between landbirds and waterbirds overlap extensively within the Mississippi Alluvial Valley. Nevertheless, waterbird goals, objectives, and priorities fall under the purview of other bird conservation groups such as the Western Hemispheric Shorebird Reserve Network, the Colonial Waterbird Society, and North American Waterfowl Management Plan Committee. The objectives outlined here will be integrated with objectives established by other conservation groups (e.g., Lower Mississippi Valley Joint Venture Evaluation Plan [Loesch et al. 1994]).

Forestry in the southeastern U.S. has a long history, with the present forested acreage referred to as the "South's Fourth Forest" (USDA Forest Service 1988, Walker 1991). Virtually no virgin timber remains and, even with increasing region-wide forest acreage, there have been dramatic changes in forest types dominating the landscape. Indeed, a number of "old-growth" forest species are now extinct, endangered, or vulnerable. Some avian species dependent upon forested habitats are extinct (e.g., passenger pigeon [Ectopistes migratorius], Carolina parakeet [Conuropsis carolinensis]) and other species are endangered and likely extinct (e.g., Ivory-billed Woodpecker, Bachman's Warbler). Non-avian species, such as the red wolf (Canis rufus) and the Louisiana black bear (Ursus americanus luteolus) have also suffered from loss of forested wetland habitat.

The Mississippi Alluvial Valley has the dubious distinction of being the most deforested of all southeastern physiographic areas. Forest fragmentation usually refers to a landscape where a still large percentage of forest remains but is fragmented into small blocks surrounded by non-forested habitat (usually agriculture or suburban/urban development). The Mississippi Alluvial Valley, however, is now largely non-forested with >80% of the land in agricultural production. Remaining forest patches are generally surrounded by vast expanses of farmland. Continuing debates on the decline of forest birds in other parts of eastern North America, where the percent of the landscape in forest remains high, have been resolved in the Mississippi Alluvial Valley. Forests, and the birds dependent upon them, have greatly declined throughout this century, with losses continuing since the 1950's (about 360,000 ha [900,000 acres] lost between the mid-1970's and mid-1980's; Hefner et al. 1995). Nevertheless, the remaining forests, of all types, provide valuable wildlife habitat.

Table 2. Hydrologic units within the Mississippi Alluvial Valley.
State	Hydrologic Unit
Arkansas	Black / Middle White, Cache / Lower White, St. Francis / L'Anguille, Lower Arkansas / Bayou Meto, Ouachita, Boeuf / Bayou Bartholomew
Illinois	Cache
Kentucky	West Kentucky
Louisiana	Boeuf / Ouachita, Tensas / Bayou Macon, Red River Backwater, Teche / Vermilion, Atchafalaya, Barataria / Terrebonne
Mississippi	Steele Bayou, Big Sunflower, Tallahatchie, Lower Yazoo, Southwest Mississippi
Missouri	Bootheel
Tennessee	West Tennessee

Conservation opportunities:

The greatest opportunity for bird conservation in the Mississippi Alluvial Valley is the existence of the Lower Mississippi Valley Joint Venture (LMVJV) of the North American Waterfowl Management Plan (NAWMP). This effort was originally devoted to betterment of conditions for waterfowl, but this infers improvement of wetland habitat and improved health of wetland ecosystems. Given that the vast majority of the MAV consists of wetlands, efforts to benefit waterfowl populations and those for forest-breeding non-game birds are often compatible or even identical in nature. Due in part to this overlap, the LMVJV has worked closely with Partners in Flight and others to develop a cohesive effort to conserve all birds within the system. The strength and success of PIF in this physiographic area is in large part due to this close partnership with the Joint Venture.

At the heart of the biological foundation of bird conservation in the MAV is adherence to roles and relationships among planning, implementation, and evaluation. The Joint Venture and its many partners have a long history of implementation during which numerous properties have been acquired by public agencies and managed for waterfowl, other game species, and/or as functional bottomland hardwood systems. Much of the currently forested land in the Valley is either on a National Wildlife Refuge, state Wildlife Management Area, National Forest, or other public property. In recent years, planning and evaluation have become much more significant components of the conservation process in the MAV, and this Bird Conservation Plan is one manifestation of this shift.

One important tool developed for planning and evaluation has been a Geographic Information System (GIS) that is now centered in the Joint Venture office in Vicksburg, MS, with significant support from The Nature Conservancy of Louisiana and other partners. This GIS has been used to characterize and map all forested habitat in the physiographic area, which has been an essential tool in development of forest patch objectives for forest-breeding birds. Public lands and other important features have also been digitized, and current work is dedicated toward mapping of flood events of various intensities in an effort to locate lands most amenable to restoration.

Another key wildlife conservation planning process in the MAV is underway for the Louisiana black bear, which is listed as federally threatened and the focus of activity of the Black Bear Conservation Committee. Like Partners in Flight, the Black Bear Conservation Committee includes representation from public agencies, private landowners, non-governmental conservation organizations, and academia; it has particularly strong representation from the forest products industry. The goal of the Black Bear Conservation Committee is to restore healthy bear populations in the Alluvial Valley and beyond. A single healthy population requires approximately 100,000 acres of bottomland hardwood forest, under some variety of management regimes. This is comparable, both in size and habitat conditions, to the larger blocks developed as forest bird objectives. The Black Bear Conservation Committee will probably recommend that five or so blocks of this size be restored or maintained between the latitude of the lower stretches of the Arkansas and White Rivers and the Gulf of Mexico. This is entirely compatible with Partners in Flight recommendations, and the combined efforts of the two groups should make this shared goal more achievable.

Management of forests for timber production is in most ways compatible with the conservation of high priority birds. Patch cuts are a widely employed silvicultural options that, within a largely forested landscape, are consistent with the habitat needs of species such as kites that forage over open areas, as well as with those of species such as the Swainson's Warbler that favor a mosaic of forest understory. Special attention to forest openings is necessary to minimize parasitism and predation even in the larger forest patches. For example, it may be advisable to maintain no more than 10-15% of an area in right-of-ways, other permanent openings, regeneration cuts, or other temporary openings at any given time. Edges within the "interior" of forests should be gradual rather than abrupt (Suarez et al. 1997). Additionally, retention of scattered patches of tall, >25 m, trees (e.g., baldcypress and cottonwood) in existing forests will provide potential nest site locations.

As another example, uneven age silviculture appears to be compatible with habitat requirements of Cerulean Warblers. However, if timber volume targets are high (e.g. >40% of the canopy removed) this type of uneven-aged management may result in excessive fragmentation. In these cases, larger regeneration cuts (e.g., 8-16 ha) combined with longer rotations (e.g., 150-200 years) and thinning to ensure maximum tree girth may still result on maintenance of large mature hardwood stands over time. Only monitoring, coupled with adaptive management, can confirm which silvicultural practices are beneficial to high priority bird species.

The importance of private lands for conservation of avian populations in the Mississippi Alluvial Valley must not be underestimated. Education of and support by private landowners is critical in that most future reforestation efforts are likely to occur on private lands. Opportunities to improve habitat conditions for avian species must consider the management objectives of private landowners. Many of the opportunities to work with private landowners to restore and manage forested wetlands through a variety private lands assistance programs are summarized in Coreil et al. 1997 and can be seen at http://www.agctr.lsu.edu/wwwac. As an example of the success of incentive programs on private lands, implementation of the USDA's Wetland Reserve Program has resulted in reforestation of over 100,000 ha of cleared land.

Probably because so much original forest has been removed from the MAV, the forest that remains is highly valued by local residents, both for hunting and aesthetic reasons. Unlike previous decades in which conversion from forest to agricultural use increased land value, most sites that are currently forested lose value when cleared. This is partially because most remaining forested sites are wetter areas least favorable to agriculture, but also because of the value of land on which hunting for deer, waterfowl, turkey, and other small game is an option. Most non-industrial private forested land is now either owned by hunting clubs or is leased for hunting to generate income. Hunting clubs are now often able to bid more for forested land than are farmers intent on conversion. Land in the ownership of limited partnership hunting clubs is extremely unlikely to be cleared for other uses anytime in the near future. The benefits of this condition to non-game forest birds are great, in that bird conservation objectives are being met on private lands, voluntarily by the landowners, at no acquisition or incentives cost to public agencies.

Long term bird conservation in the Mississippi Alluvial Valley can best be sustained through implementation of detailed planning, communication, and working partnerships among an array of government agencies, private conservation organizations, landowners, and citizens. This process will lead to an understanding of this region's natural history, foster regional pride, and instills the conservation ethic necessary to sustain and restore the diverse biotic communities of this ecosystem. Various groups exist that can merge federal, state, and private interests to coordinate management actions and resolve potential conflicts. For instance, much of the current conservation planning effort underway in the Mississippi Alluvial Valley was outlined through the efforts of the Lower Mississippi Valley Joint Venture sanctioned under the North American Waterfowl Management Plan. However, much work needs to be done to better coordinate management needs with the private sector. Landowner incentives, conservation easements, and market development can all be used to increase private participation in conservation efforts.

Within the Mississippi Alluvial Valley, data on land cover (U.S. Geological Survey, Twedt 1996) and ownership (U.S. Fish and Wildlife Service, The Nature Conservancy), forest seral stages, and trends (U.S. Forest Service [www.srsfia.usfs.msstate.edu] are available for assessment of avian habitat. State Working Groups of Partners in Flight can use these data, and the recommendation herein, to assess their public and private land opportunities, thereby defining local roles and responsibilities for achieving the population goals and habitat objectives outlined below.

SECTION 2: AVIFAUNAL ANALYSIS

Species prioritization

At least 107 bird species nest regularly in the Mississippi Alluvial Valley, excluding wading birds and colonial nesting waterbirds (Appendix 2). Most of these species occur in more than one broad habitat type as defined within this plan. Forest breeding species remain the most important component of the avifauna, despite the loss of nearly 80% of the forested wetlands in this region. At least 70 species occur in bottomland hardwoods as a primary habitat. Greater than 20% of the breeding populations of Swainson's Warbler, Prothonotary Warbler, and Swallow-tailed Kite are found within the Mississippi Alluvial Valley. Typical species of bottomland hardwood forests include Northern Parula Warbler, Swainson's Warbler, Prothonotary Warbler, Red-shouldered Hawk, and Red-headed Woodpecker.

At least 62 species occur in upland oak-hickory forests, although many of these species occur in bottomland hardwoods as well. Typical species include Yellow-billed Cuckoo, Worm-eating Warbler, Black-and-white Warbler, and Broad-winged Hawk. At least 25 species occur in scrub-shrub habitats in the Mississippi Alluvial Valley. Typical species include Painted Bunting, Orchard Oriole, White-eyed Vireo, Common Yellowthroat, and Indigo Bunting.

At least 18 species occur in Mississippi Alluvial Valley grassland habitats. Typical species include Dickcissel, Loggerhead Shrike, Field Sparrow, Northern Bobwhite, and Grasshopper Sparrow. Seven species are dependent on water and wetlands habitat, including Hooded Merganser, Wood Duck, Belted Kingfisher, and Marsh Wren. Another 19 species occur in a variety of other habitats, such as open water or river banks. These species include Purple Martin, Tree Swallow, and Barn Owl.

The Partners in Flight prioritization process was developed to guide conservation actions among diverse birds and habitats (see Hunter et al. 1993, Carter et al. in press). The system ranks each species based on 7 measures of conservation vulnerability: relative abundance, size of breeding and non-breeding ranges, threats during breeding and non-breeding seasons, population trend, and relative density. In addition, the percentage of a species global breeding population that occurs in a physiographic area has been provided (Rosenberg and Wells, pers. comm.). To further refine species prioritization within a physiographic area, population trends and area importance are examined independently of total scores.

Birds were prioritized according to this scheme in the Mississippi Alluvial Valley (Table 3). Category I lists highest priority birds and included 14 species which received a Partners in Flight score of 22 or more. Two species were excluded from the prioritization process in the Mississippi Alluvial Valley, Bachman's Warbler and Ivory-billed Woodpecker are most likely extinct. Other species of highest priority are Swainson's Warbler, Prothonotary Warbler, Cerulean Warbler, and Swallow-tailed Kite. The remainder of Category I birds occur mostly in bottomland hardwood forests, although birds of upland forests and scrub-shrub habitats are represented.

Category II provides a list of slightly lower priority species, and included another 9 species with slightly lower total scores (19 to 21), but combined with a high score for area importance and population trend. These species reflected mostly scrub-shrub habitats. Four species are included in Category III. These birds received high global concern scores as Watchlist species (Cater et al. in press), regardless of their status in the Mississippi Alluvial Valley.

Category IV birds have a high combination of scores for area importance and population trend, regardless of total score. In the Mississippi Alluvial Valley, 4 species occur in this category. Nine species are in Category V; each species has greater than 5% of their national breeding status within the Mississippi Alluvial Valley. Species with greater than 10% of their breeding population, that have not been listed elsewhere in this table, include Barred Owl and

Mississippi Kite. Categories VI and VII list federally threatened and endangered species, as well as species of state concern. Combined, these categories include 7 species.

Table 3. Priority bird species listed by total PIF concern score, and segregated by entry criteria. Other measures include are of importance and population trends scores, percent of BBS populations, and local migratory status. [Refer to Appendix 2 for scientific names]
Priority and species	Total PIF score		Concern scores A1 PT		Percent BBS			Local migratory status
Ia. Highest overall priority
Swainson's Warbler		29	5	3	20.8			B
Swallow-tailed Kite		28	4	3	25.1			E
Cerulean Warbler		28	3	4	-			E
Ib. High overall priority
Prothonotary Warbler		24	5	2	34.8		B
Painted Bunting		24	3	5	4.4		B
Red-headed Woodpecker		23	5	5	3.0		D
Bell's Vireo		23	2	3	1.0		B
Northern Parula		23	5	5	6.9		B
Worm-eating Warbler		23	2	3	-		B
Kentucky Warbler		22	3	3	4.7		B
Orchard Oriole		22	5	5	7.4		B
Yellow-billed Cuckoo		22	5	5	6.0		B
Wood Thrush		22	3	3	1.3		B
White-eyed Vireo		22	4	5	8.4		B
II. Physiographic area priority species
Yellow-breasted Chat		21	5	5		6.2	B
Northern Bobwhite		20	3	5		-	R
Eastern Wood-Pewee		20	3	5		-	B
Carolina Chickadee		20	4	5		-	R
Loggerhead Shrike		20	4	4		-	R
Field Sparrow		20	3	5		-	D
Baltimore Oriole		20	3	5		-	B
Ruby-throated Hummingbird		19	5	3		7.3	B
Blue-gray Gnatcatcher		19	4	5		-	B
III. Additional species: global priority
Scissor-tailed Flycatcher		21	3	3		-	B
Dickcissel		21	4	2		5.1	B
Chuck-will's-widow		21	4	3		3.1	B
Prairie Warbler		20	2	3		-	B
IV. Additional species: abundant and declining in physiographic area
Indigo Bunting		17	4	5		-	B
Common Grackle		16	5	5		-	D
Mourning Dove		14	4	5		-	D
Northern Mockingbird		14	4	5		-	D
V. Additional species: responsibility for monitoring (>5% BBS population estimate)
Mississippi Kite		21	4	2		13.4	B
Acadian Flycatcher		20	3	2		5.6	B
Carolina Wren		18	5	3		6.5	R
Red-bellied Woodpecker		18	5	2		6.1	R
Red-shouldered Hawk		17	4	2		9.8	D
Purple Martin		17	5	2		7.8	B
Barred Owl		16	5	2		15.6	R
Northern Cardinal		16	5	2		5.7	R
Black Vulture		12	3	3		8.3	D
VI. Federal listed species
Bald Eagle		18	3	3		-	D
VII. Local, state, or regional interest species
Hooded Warbler		21	3	3		-
Yellow-throated Warbler		20	3	2		-
American Redstart		20	3	3		-
Yellow-throated Vireo		20	3	2		-
Summer Tanager		18	2	3		-
Pileated Woodpecker		16	4	2		-

Conservation area size considerations

Other than the above issues of site characteristics, the basic consideration regarding habitat is that the current landscape of the MAV is 80% deforested, most remaining forested patches are small and isolated (Twedt and Loesch, in press), and there is no reason to presume that the vast majority of this system will ever again be forested. Objectives for bird populations in forested habitat, therefore, are related to size, configuration, number (and to some extent distribution and inter-connectivity) of individual habitat patches. The quantity of habitat required can conceptually be separated into

issues of 1) patch size and context; and 2) number and distribution of patches (populations). Size and configuration will be addressed in this section whereas the issue of number of patches and their distribution will be addressed thereafter.

Forest patches should be of sufficient size to support source populations of targeted bird species, to minimize the likelihood of extirpation, and to ensure a low probability of genetic degradation. The issues that must be resolved in order to select an appropriately-sized patch of habitat for breeding birds are: 1) context, or how ought breeding birds be buffered from the negative influence of surrounding matrices; 2) desired number of breeding pairs to constitute a source population with a high probability of long-term viability; and 3) the density at which birds tend to occur within habitat likely to be included in average patches. Area can be conceptually calculated using the formula:

A = (N * D) + B,

where A = Area of forest required to support a source population (ha),

N = Desired number of breeding pairs

D = Density of breeding birds (expressed as area / breeding pair), and

B = The area (ha) of a 1-km-wide forested buffer around the core forest area (N * D).

The agricultural matrix in which forest patches in the MAV are embedded is generally considered hostile to forest breeding birds in that many MAV land uses support Brown-headed Cowbirds and a wide range of nest predators. We assume that the edges of forest blocks are more seriously affected by brood parasitism and nest predation than are areas farther into the interior of blocks, although the relationship between negative effects distance from edge has not been rigorously tested in the MAV. However, there are data that indicate that the distance between breeding and feeding locations for Brown-headed Cowbirds averages 1.2 km (Thompson 1994). For planning purposes, therefore, we have assumed that the negative impacts on a core surrounded by a 1 km forest buffer will be reduced. Furthermore, for at least some species, birds breeding in forest interiors are more likely to be paired with a mate than are birds breeding near forest edges (Van Horn et al. 1995). Only those pairs within the forest core, therefore, are assumed to reproduce at a rate sufficient to serve as a source population. Clearly, the assumptions in this logical process need to be tested.

The area occupied by a 1-km-wide buffer will vary with the geometric configuration of forest patches. For planning purposes, until the actual area of interior forest within each forest patch is determined, doubling the core forest area (i.e., 2 *[N * D]) will generally result in a total area that includes a 1-km-wide buffer around the desired interior forest area. This assumption holds for more compact small forest patches. For example, a relatively compact 4,000 ha forest patch would be expected to have 2,000 ha of forest interior inside its 1-km-wide forest buffer. This interior area is assumed capable of harboring 500 breeding pairs of all bottomland forest species that occur at densities of <4 ha per breeding pair (i.e., 500 pairs * 4 ha/pair = 2000 ha). The ratio of land required to maintain a 1 km wide buffer to interior core area increases as block configuration becomes less compact.

Minimum viable population and source populations

The concepts of minimum viable populations and source populations are relevant to considerations of desired size, but, unfortunately, what constitutes these threshold population sizes remains unknown for almost all species. Minimum viable population sizes have been calculated from 250 (Reed et al. 1988) to several thousand (Thomas 1990). A somewhat arbitrary population goal of 500 breeding pairs (1000 breeding individuals) per forest patch was adopted for the MAV. This goal raises a number of concerns, but 500 pairs sufficiently buffered from strong negative influences on reproductive success is assumed to be large enough to constitute a source that is unlikely to undergo extirpation in a patch as long as habitat quality is maintained.

Area recommendations are based upon the amount of buffered forest interior core habitat necessary to support 500 breeding pairs of the high priority birds that occur at lower densities than do other priority species. Assuming that microhabitat needs are satisfied, an area that supports 500 pairs of the least densely distributed species ought to support 500 or more pairs of all of the species that occur at greater densities. It is recognized that at least some of these density data used were derived from study sites with optimal habitat conditions, and that large-scale habitat restoration will be accomplished in a variable landscape, often including sites unlikely to be able to support densities reported in the literature. These data and assumptions must be improved upon in the future.

Based on density estimates in Hamel 1992b and empirical observations (Table 4; Mueller et al., in press), there are three high priority birds that each can be used to define area thresholds: Swainson's Warbler, Cerulean Warbler, and Swallow-tailed Kite. Based on area (and to some extent habitat quality) needs, each of these three birds can be a "focal" or "umbrella" species for each of three general size thresholds.

The Swainson's Warbler occurs throughout the area; viable source populations presumably can persist in patches equaling or exceeding 4000 ha in size (assuming that configuration allows about half of that to be protected by a 1 km wide buffer). Other birds expected to do well in this sized patch include Blue-gray Gnatcatcher and American Redstart. The next sized patch, set at about 8000 hectares, supports Cerulean Warblers (but only in the northern half of the valley where they breed) as well as Yellow-throated Warbler, Kentucky Warbler, Yellow-billed Cuckoo, Eastern Wood-pewee, Great Crested Flycatcher, Scarlet Tanager, Summer Tanager, and Yellow-throated Vireo (all of which occur throughout the area). The threshold of the largest sized patch, 40,000 ha, supports Swallow-tailed Kites, but as far as is known in numbers well below 500 pairs. Birds for which such sites could serve as source areas include Cooper's Hawk, Red-shouldered Hawk, and Broad-winged Hawk. Additionally, source populations of some species should thrive in forest patches smaller than any identified in this plan, such as those that contain interior areas of between 2,000 to 4,000 ha. These species include Prothonotary and Hooded Warbler, Northern Parula, Wood Thrush, Acadian Flycatcher, and Red-eyed Vireo.

The step taken prior to setting population/patch number objectives was to assess the current status of suitable habitat patches. All forested habitat was classified from satellite imagery. Maps produced through this process have been invaluable tools in all phases of bird conservation planning. All forest patches were categorized as <4,000 ha, 4-8,000 ha, 8-40,000 ha, and >40,000 ha. Each patch was then further assessed to determine whether restoration of additional forested area could move the patch into the next higher category or whether two or more patches could be joined to form a larger patch.

Swainson's Warbler and Prothonotary Warbler are among the most highly ranked species within the Mississippi Alluvial Valley but their habitat and area requirements are perhaps more easily meet than are those of other highly rank species. A source population of Swainson's Warbler probably requires >2500 ha (4700 ha in agriculturally-dominated landscapes) of mature forested wetlands, whereas Prothonotary Warblers probably require only 1600 ha (2700 ha in agriculturally-dominated landscapes) of bottomland forest. Source populations of the remaining highly ranked species all require >7000 ha of forested habitat (Table 4).

Table 4. Hypothesized forest area (hectares) required to support viable populations of 500 breeding birds within the Mississippi Alluvial Valley (from Mueller et al. 1999). [Refer to Appendix 2 for scientific names]
Species	Patch Size Recommendation	Habitat Area Objective
Swainson's Warbler	4,700	4,000
Cerulean Warbler	8,000	8,000
Swallow-tailed Kite a	40,000	40,000
Prothonotary Warbler	2,700	4,000
Northern Parula	3,000	4,000
Hooded Warbler	2,500	4,000
Kentucky Warbler	8,400	8,000
Yellow-billed Cuckoo	6,600	8,000
Wood Thrush	2,800	4,000
Louisiana Waterthrush	7,200	8,000
Acadian Flycatcher	2,800	4,000
Eastern Wood-pewee	5,500	8,000
Yellow-throated Vireo	7,900	8,000
Yellow-throated Warbler	7,800	8,000
Blue-gray Gnatcatcher	4,000	4,000
Summer Tanager	6,600	8,000
Great-crested Flycatcher	7,200	8,000
Red-shouldered Hawk	57,800	40,000
Scarlet Tanager	4,900	8,000
Red-eyed Vireo	1,800	4,000
American Redstart	4,600	4,000
Broad-winged Hawk	101,000	40,000
Pileated Woodpecker	19,000	40,000
Cooper's Hawk	45,000	40,000
White-breasted Nuthatch	8,600	8,000
^aBased on Cely and Sorrow (1990), a 40,000 ha patch of bottomland hardwood forest would only support approximately 80 pairs of Swallow-tailed Kites. A secure (source) population would realistically have to be based on a regional (southeastern U.S.) population.

SECTION 3: HABITATS AND OBJECTIVES

Breeding birds are grouped into 6 priority habitat-species suites for the Mississippi Alluvial Valley (Table 5). Each of these habitats, forested wetlands, forest openings and scrub-shrub, upland oak-hickory, grasslands and savannas, nonforested wetlands, and urban, suburban, rural woodlots, is discussed below.

Table 5. Area of bottomland hardwood forest types within the Arkansas, Louisiana, and Mississippi portion of the Mississippi Alluvial Valley. Data from McWilliams and Rosson (1990). Society of American Foresters (SAF) cover type from Eyre (1980). Temporarily Flooded Forest Alliance (TFFA), Seasonally Flooded Forest Alliance (SFFA), and Semipermanently Flooded Forest Alliance (SPFFA) name and identification number from The Nature Conservancy (1997). [Refer to Appendix 1 for scientific names]
Forest Type	Hectares
Swamp chestnut oak - cherrybark oak (SAF cover type 91) I.B.2.N.d.210. (Swamp chestnut oak, cherrybark oak, Shumard oak) - sweetgum TFFA	30,000
Cottonwood (SAF cover type 63) I.B.2.N.d.160. cottonwood TFFA	52,000
Sweetgum - Nuttall oak - willow oak (SAF cover types 88, 88 var., 92) I.B.2.N.d.250. willow oak, water oak, diamondleaf oak TFFA I.B.2.N.e.130. willow oak SFFA	433,000
sugarberry - American elm - green ash (SAF cover type 93) I.B.2.N.d.090. sugarberry - cedar elm TFFA I.B.2.N.d.110. green ash - American elm TFFA I.B.2.N.d.260. Nuttall oak - sugarberry - (American elm, cedar elm) - (honey locust) TFFA	534,000
willow (SAF cover type 95) I.B.2.N.d.280. black willow TFFA I.B.2.N.e.160. black willow SFFA	163,000
overcup oak-water hickory (SAF cover type 96) I.B.2.N.e.032. water hickory - water locust SFFA I.B.2.N.e.100. overcup oak - water hickory SFFA I.B.2.N.e.140. Nuttall oak - (overcup oak) SFFA	248,000
cypress-tupelo (SAF cover types 101, 102, 103) I.B.2.N.d.290. bald cypress - sycamore TFFA I.B.2.N.e.190. bald cypress - (water tupelo, swamp blackgum, ogeechee tupelo) SFFA I.B.2.N.f.030. bald cypress - (water tupelo, swamp blackgum, ogeechee tupelo) SPFFA	396,000
Other forest types sycamore - sweetgum - American elm (SAF cover type 94) live oak (SAF cover type 89) river birch - sycamore (SAF cover type 61) silver maple - American elm (SAF cover type 62) pin oak - sweetgum (SAF cover type 65) I.B.2.N.d.010. box elder TFFA I.B.2.N.d.030. silver maple TFFA I.B.2.N.d.040. sugar maple - bitternut hickory TFFA I.B.2.N.d.050. river birch - (sycamore) TFFA I.B.2.N.d.070. pecan - (sugarberry) TFFA I.B.2.N.d.140. sycamore - (green ash, sugarberry, silver maple) TFFA I.B.2.N.d.200. bur oak - swamp white oak - (kingnut hickory) TFFA I.B.2.N.d.255. willow oak - cedar elm TFFA I.B.2.N.e.020. red maple - green ash SFFA I.B.2.N.e.045. (eastern mayhaw, western mayhaw, rufus mayhaw) SFFA I.B.2.N.e.090. planertree SFFA I.B.2.N.e.120. pin oak - (swamp white oak) SFFA	84,000
Total bottomland hardwood	1994,000

Forested wetlands, bottomland hardwood forests

Status and importance

Bottomland forests, floodplain forests, and swamps in the Mississippi Alluvial Valley are adapted to and a result of frequent floods. High tree species diversity, dense midstories and understories, and various levels of wetness create conditions that support many species not commonly found away from forested wetlands. The most frequently occurring wetland forest

cover types making up the oak-gum-cypress forest complex within the Mississippi Alluvial Valley are sugarberry-American elm-green ash, sweetgum-Nuttall oak-willow oak, and cypress-tupelo (Table 6). Understory development within each of these forest types is dependent on local flooding patterns and canopy density.

Extensive drainage and conversion of forested wetlands within the Mississippi Alluvial Valley has increased the extent of farmland, pine or hardwood monocultures, and residential, industrial and commercial development (Sharitz and Mitsch 1993). Drainage and clearing of floodplain forests was underway by the mid-1800's and by the 1940's less than 5 million ha (12 million acres) remained. Significant forest clearing occurred in the 1940's and from the 1960's to the early 1970's (McWilliams and Rosson 1990). By 1978, the 2.1 million ha that remained were the wettest areas, consisting of 60% seasonally flooded forests and 40% swamps (MacDonald et al. 1979). Although forested wetland losses have continued, the rate of loss has slowed (McWilliams and Rosson 1990). The remaining forested wetlands in the Mississippi Alluvial Valley have almost all been cut over at least once. What remains forested is highly fragmented. Of the more than 38,000 forest patches >2 ha within the Mississippi Alluvial Valley, only 267 patches were over 1,012 ha but they contained >50% of the total forest area. The average patch size is 64 ha (USGS, unpublished data).

In forested wetlands in the MAV, habitat quality can be limited to issues of vegetative composition and structure, which are determined by moisture regimes and disturbance patterns. Due to a ridge and swale topography, most sites of any size will include low, wet areas as well as some that are higher and drier. Most large sites will therefore include a diversity of conditions that in sum are potentially satisfactory for all of the birds in this species suite. This microhabitat diversity is important because high priority species dependent upon forested wetlands vary in their microhabitat requirements (Pashley and Barrow 1992). Because the driest sites are best for agriculture, however, places that remain in forest tend to represent the wettest conditions in the Valley. The primary issue of concern regarding quality is that a sufficient amount of dry ridge habitat be included in each patch or set of patches to support those birds (including Swainson's Warbler and Cerulean Warbler) that prefer the structure associated with dry sites.

Disturbance patterns other than flooding also affect habitat quality. Historically these were treefalls caused by wind, ice storms, or other climatic events or perhaps very infrequent fire in the drier areas. The immediate result of each of these disturbances was a small to moderate-sized gap into which sunlight penetrated that was quickly occupied by shrubs, vines, and saplings. These gaps may be used preferentially by some birds (perhaps Hooded Warblers and surely White-eyed Vireos). The longer-term result was that each gap grew up as a small even-aged group of trees, resulting in a forest that consisted of a mosaic of many such even-aged groups of varying ages. An "old growth" bottomland hardwood, therefore, was not a continuous closed canopy that we associate with many older forests, but rather a dynamic mixture of patches ranging in age from very young to very old (Pashley and Barrow 1992, Keith Ouchley, pers. comm.).

Table 6. Habitat objectives for predominately mature, forested wetlands within the Mississippi Alluvial Valley. [See also Fig.2 and Appendix 5]
State	Number of Bird Conservation Areas	Number of forest patches (ha)
		4,000-8,000	8,000-40,000	>40,000
Arkansas	18	9	11	3
Illinois	1	0	1	0
Kentucky	4	3	1	0
Louisiana	33	19	15	7
Mississippi	21	14	6	2
Missouri	7	6	1	0
Tennessee	3	1	1	1
Totals	87	52	36	13

Priority species, species suites, and habitat requirements

Swainson's Warbler habitat has been described as "rich damp woods with deep shade, moderately dense undergrowth, and relatively dry ground" (Meanly 1971). Swainson's Warblers favor breeding sites that are moist but not flooded with understory vegetation of cane (Arundinaria gigantea), dwarf palmetto (Sabal minor), and sweet pepperbush (Clethra alnifolia) (Meanley 1971, Eddleman et al. 1980). These habitats may be produced in forested wetlands where tree fall or selective harvest (thinning, single tree selection, or patch cuts) of closed canopies has allowed light to hit the ground, in turn allowing vegetative development. Many of these sites are now prime agricultural sites for cotton production due to edaphic and hydrologic features that favor better drainage. Thus, the sites most conducive to restoration (less productive cropland and frequently flooded areas) are often not prime Swainson's Warbler habitat. However, because of topographic variability, larger restoration sites are likely to include high ground that supports Swainson's Warblers. Populations of this species at the northern end of the Mississippi Alluvial Valley (north of the Tensas Basin) are increasingly isolated from populations in the coastal plains and the southern Mississippi Alluvial Valley. Birds in small forest patches, as in Missouri, may suffer from effects of genetic isolation (Thomas et al.1996 ; G. Graves, Smithsonian Institution, pers. comm.).

Generally numerous wherever moderately large forested wetlands are flooded through the breeding season, Prothonotary Warblers are probably the most secure of the high priority species within the Mississippi Alluvial Valley. This is a result of comparatively small area requirements and preference for the flooded forested habitat that persists in the Mississippi Alluvial Valley. However, nest depredation and parasitism rates may be high for this cavity-nesting species and should be monitored where feasible.

The only currently known Swallow-tailed Kite breeding populations in the Mississippi Alluvial Valley are in the Atchafalaya Basin of Louisiana. This species historically bred in 21 states with concentrations in 9, but now breeds in only 7 states with concentrations only in Florida (Meyer 1990, 1995; Meyer and Collopy 1990). Since the turn of the century, it has suffered the most dramatic reduction of any extant landbird in eastern North America.

The North American subspecies of Swallow-tailed Kite is probably the most vulnerable bird in the southeastern U.S. that is not federally listed as a threatened or endangered species nor a candidate for such listing. Total population size for this species is unknown but is certainly <5000 individuals, and probably <1150 breeding pairs (Meyer and Collopy 1990). This species appears to be expanding into some river systems where it has been absent from most of this century (e.g., Sabine, Louisiana-Texas, and Pee Dee, South Carolina). This suggests that the population may expand in response to improved habitat conditions. About 40,000 ha (100,000 acres) of mature forested wetlands appear to be necessary to support 80 to 100 kite pairs (Cely and Sorrow 1990). Within patches of sufficient size for Swallow-tailed Kites, management should increase the availability of nest sites (i.e., small stands of >30 m (100 ft) tall or higher trees, presumably cottonwood and baldcypress). Security for the Swallow-tailed Kite may require maintenance of >80 breeding pairs in 8 of the 13 proposed forest patches of >40,000 ha in the Mississippi Alluvial Valley in combination with populations in 13 major coastal plain floodplains from South Carolina to Texas outside of peninsular Florida. Mississippi Alluvial Valley forested wetlands may serve as staging sites from which Swallow-tailed Kites could reclaim their former breeding distribution throughout the mid-western United States.

Within the Mississippi Alluvial Valley, Cerulean Warblers breed almost exclusively north of the Arkansas-Louisiana state boundary. Although isolated breeding pairs have been reported in north Louisiana and in Mississippi, no significant populations are currently known from either of these states. Although persisting in some numbers in the highlands and plateaus from the southern Appalachians westward, this species has been much reduced within its historical breeding distribution throughout the southeastern coastal plain, including the Mississippi Alluvial Valley. Hamel (1992a; also see Robbins et al. 1992) surmised that a minimum of 4,000 ha of mature forested wetlands was required to maintain a source population of Cerulean Warblers based on his work in Tennessee. In areas where the landscape is dominated by agriculture, however, a more conservative estimate of 8,000 ha may be necessary to support source populations. Past forestry practices, hydrology, and perhaps other factors confound our ability to predict the locations of Cerulean Warbler populations. Habitats occupied by Cerulean Warblers tend to contain numerous very large trees towering above a multi-layered canopy structure and interrupted by frequent canopy gaps (Hamel 1992a, Robbins et al. 1992). It is important to note that Cerulean Warblers persist in commercial forests in which timber harvests mimic tree fall gaps (Hamel 1992a).

Although this plan only explicitly addresses the forested wetland needs of breeding birds, this physiographic area also provides important wintering habitat for temperate migrants and in-transit habitat for long-distance migrants. We have tentatively assumed that the habitats provided for breeding birds are also sufficient for wintering and migrant birds. This assumption needs to be rigorously tested. Additionally, some areas not suitable for high priority breeding birds may be very important for wintering and transient birds.

Habitat and population objectives

Population goals in the MAV are to ensure maintenance of conditions suitable for support of source populations of target species in 52 predominately mature, forested wetlands patches >4,000 ha (>10,000 acres), 36 patches >8,000 ha (>20,000 acres), and 13 patches >40,000 ha (>100,000 acres). However, other forest patches that are smaller than the minimum required to support source populations may still benefit breeding, wintering, or migrating birds. The distribution of patches among states (Table 7, Appendices 4 and 5) is not even, largely reflecting existing opportunities. For example, there are more and larger blocks in south Louisiana than in Missouri. In order to improve the distribution of blocks, more marginal blocks were included in under represented areas. Patch size objectives were generally set at the highest threshold of realistically achievable restoration (Fig. 2) and distributed among states. This was a practical assessment of what is relatively easy to achieve. The main remaining question is whether it is enough. At least two relevant points to consider in answering this question are population viability and percentage of ecosystem health being maintained.

Although he strongly cautioned that minimum viable population size should be independently calculated for each species, Soulé (1987) suggested that a population size "in the low thousands" should be adequate for most vertebrates. Individuals breeding within a forest patch, however, are probably not genetically isolated from individuals in other patches. Because virtually all species of high concern in the MAV are neotropical migrants that exhibit low natal site fidelity, there is presumably a high likelihood of gene flow among bird populations breeding in different forest patches. Thus, maintaining avian populations above the "low thousands" in the entire physiographic area may assure viability of most species from a genetic perspective. The possible exception to confidence in this regard is the Swallow-tailed Kite, for which it may be necessary to perform a population and habitat viability analysis (Beardmore, in press) throughout the species' range in the Southeast to generate reliable population goals.

There is a level of effectiveness below which ecosystems begin to lose species. Effectiveness includes some unknown combination of extent and function, and the threshold below which loss occurs is also largely unknown. Even though much about this is unknown, it is safe to say that the Mississippi Alluvial Valley bottomland hardwood ecosystem, through some combination of reduction in acreage, fragmentation, change in age and structure, and/or hydrological alteration, has fallen below that level. Ivory-billed Woodpecker and Bachman's Warbler are extinct, the red wolf is extirpated, and black bears and Swallow-tailed Kites are relegated to tiny segments of their former range.

It has been suggested that at least 10% of an ecosystem be maintained to prevent extirpation of species (Jon Haufler, pers. comm.). Even though something like 20% of the MAV remains forested, the vast majority of that forest exists as tiny fragments in which all but the least area-sensitive wildlife cannot persist. A reasonable objective to this bird conservation plan, therefore, could be to restore or maintain 10% of the original area of the MAV as functional forest habitat.

Although this rather arbitrary objective is enticing in many ways, it also raises some logical inconsistencies. First, the system has already lost those species that the theoretical maintenance of a 10% level is supposed to retain. Those organisms next most threatened with extirpation in the MAV may be big river fish. There is no doubt that aquatic organisms will benefit from reforestation, but the degree to which they will enjoy protection as a result of restoration to health of 10% of the terrestrial system is unknown. It is likely that the status of aquatic organisms will continue to be precarious without major hydrological improvements beyond the scope of this plan.

Will implementation of the objectives of this plan bring back the most area-sensitive terrestrial organisms that remain - the Swallow-tailed Kite and Louisiana black bear? Each of these animals is postulated to require patches of 40,000 ha of forest. Regardless of the sum of the total area in this plan's objectives for the 100 forest patches, the sum of acreage in patches exceeding 40,000 ha is considerably less than 10% of the original extent of the system. A second and rather surprising logical consideration of a 10% objective, therefore, is that the 100 forest patches recommended herein are insufficient.

An objective of 100 healthy populations, including 36 that support birds with area needs of 8,000 ha and 13 for birds requiring 40,000 ha is surely enough to maintain viability of all save perhaps one species of bird. Even this may be biologically acceptable if the sum of objectives of this plan plus those of plans for nearby physiographic areas in the Southeast is enough to provide long-term security for the Swallow-tailed Kite. On the other hand, although the objective adds up to a very significant area, it is nonetheless doubtful that it is enough to recreate and perpetuate a healthy bottomland hardwood ecosystem in the MAV.

In some ways, however, the issue of sufficiency of population goals at the physiographic area level is not biological in nature, but instead depends upon our anticipation of the demands of society in future centuries for populations of birds and other elements of biological diversity. From this perspective, it is difficult to evaluate the sufficiency of these ambitious but realistic goals.

Table 7. Area in hectares of forested habitat within the Mississippi Alluvial Valley based on 1992 satellite imagery.
State	Total land area	Forested area	Swamp forests	Bottomland hardwoods	Thin-edge forest	Number of patches > 1012 ha
Arkansas	3,755,000	718,000	93,000	420,000	115,000	67
Illinois	55,000	10,000	1,000	5,000	3,000	3
Kentucky	69,000	20,000	4,000	12,000	3,000	14
Louisiana	3,835,000	1,397,000	411,000	719,000	163,000	130
Missouri	1,008,000	61,000	4,000	39,000	14,000	7
Mississippi	2,038,000	454,000	64,000	314,000	62,000	38
Tennessee	273,000	59,000	9,000	36,000	11,000	8
Total	11,033,000	2,719,000	586,000	1,545,000	371,000	267

Implementation recommendations and opportunities

Within the context of the Partners in Flight "Flight Plan", each of these blocks is a Bird Conservation Area, in that each block is intended to support a healthy source population of all targeted bird species. Site specific planning within each of these Bird Conservation Areas will be based upon size and configuration of existing forests, ownership and intent of landowners, flood regimes, and our current knowledge of the avifauna. An example of a detailed process to determine site specific objectives is provided in Appendix 6. In general, the long-term security of forest habitat on public land, private industrial forest products property, and in land held under limited partnerships (e.g., hunting clubs) is considered high. A variety of conservation tools may be employed on lands in other ownership categories. Private landowner involvement will be essential, in that land acquisition by public and private conservation agencies will never be adequate to accomplish these objectives. Subjective (Appendix 4) and objective (Appendix 5) criteria can be used to prioritize restoration efforts among proposed Bird Conservation Areas. Even so, we cannot ensure that bird population goals will be attained simply by implementing habitat strategies. Research and monitoring are required to assess progress and to refine both population goals and habitat objectives

Land use and management within Bird Conservation Areas will determine whether they are used to their fullest by high priority bird species. Within limits set by topography and location, areas in which natural hydrological processes are allowed to influence native plant communities will create conditions most suitable for most of the birds. Irreversible historical alterations and landowner objectives limit this option to only a few of the Bird Conservation Areas. In other areas, maintenance of vegetative structure preferred or required by high priority birds may be compatible with other management objectives. For example, Cerulean Warblers may benefit most from fast growing trees like cottonwoods and tulip-poplars, but current reforestation is usually concentrated on slow-growing species such as oaks and pecans. Nevertheless, reforestation emphasizing oaks and hickories can be easily modified to include soft-seeded, faster growing trees. Other forest management practices can be modified to provide denser understory vegetation for many understory breeding birds.

Modern-day disturbance, largely silvicultural activities, can mimic the historical pattern of a landscape mosaic of young and old forests. Small clear cuts or selective harvest can leave conditions similar to treefall gaps, and can result in forest consisting of a mosaic of patches of different ages, with each patch even-aged internally. Modern forestry does not, however, tend to leave patches in older age classes. The one bird most closely associated with those older conditions, the Ivory-billed Woodpecker, is now extinct, so the negative impact of such omission is not obvious. In some ways, industrial forests mimic natural conditions better than many protected areas, in that the forests now protected were last cut midway through the century, which means that each large area is even-aged and about forty to eighty years old. This is too young for a forest to begin undergoing much in the way of treefall, which results in a lack of gaps and a greatly reduced volume and diversity of understory and midstory vegetation. Given time, protected forest will assume the diverse characteristics of pre-settlement bottomland hardwoods. In the meantime, it is possible that timber stand improvement practices, single tree selection, and/or group selection could hasten the development of structure and benefit some of the bird species most dependent on mid- or understory vegetation.

All of these recommendations for forest-breeding birds must be integrated with objectives set for waterfowl and shorebirds. Bottomland hardwood forests are important habitat for waterfowl during winter, but whether the amount and distribution of forest for breeding birds is sufficient for waterfowl has yet to be rigorously examined. Some waterfowl and all shorebird needs are to be met on open wetlands. The balance between open and forested land on public property and issues of the impact of open land on adjacent forest (e.g., do these areas support large numbers of cowbirds and predators?) also needs to be investigated. Also, there must be a concerted effort, in the future, to assure that these bird conservation areas adequately represent the range of naturally occurring soil and community conditions in the MAV. Failure to meet these conditions may ultimately require more restoration and expense than currently contemplated.

Evaluation of assumption

Population objectives for forest birds are based upon the assumed spatial requirements of territorial individuals. It is assumed that habitat availability is the primary limiting factor for these birds and that maintaining or restoring "suitable" forest patches (i.e., patches of adequate size, shape, and management) will result in bird population increase and eventual stabilization. We also assumed that the breeding bird densities recorded in Hamel (1992b) reflect the average densities of birds in the Mississippi Alluvial Valley and that the target number of 500 breeding pairs of the appropriate species (with some exceptions, such as Swallow-tailed Kite) will occur within forest patches exceeding minimum size thresholds. Each forest patch that meets designated criteria is assumed to support a source population of each of its representative breeding species; that is, populations that on average produce more offspring than the number required to compensate for adult mortality. Finally, we assumed that gene flow among populations in forest patches is sufficient to maintain genetic diversity. The assumptions in this Plan will be tested on the basis of the following objectives.

Objective 1: The total area, geographic distribution, fragment size, interior area, and qualitative characteristics (e.g., ownership, habitat type, hydrological regime, etc) of bottomland hardwood forests within the Mississippi Alluvial Valley must be determined. Much of this has already been done. Forest cover within the Mississippi Alluvial Valley has been classified from thematic mapper (TM) imagery (Table 8). Discrete forest patches have been delineated and public land boundaries have been defined. Now available are data on the total area of bottomland hardwood forest by state, the number, size distribution, and geographic distribution of forest patches, and the area of forest in public and in private ownership. The Lower Mississippi Valley Joint Venture office is continuing to pursue other applications of Geographic Information Systems, including the distribution of flood stages across all habitat types. Long-term monitoring strategies are required to gauge progress toward habitat objectives and ultimately population goals.

Table 8. Area of shrub/scrub and grassland habitats (in hectares) within the Mississippi Alluvial Valley classified from 1992 satellite imagery (USGS, unpublished data).
State	Scrub/Shrub Habitats^a	Grassland Habitats^b
Arkansas	206,000	120,000
Illinois	3,000	1,000
Kentucky	5,000	3,000
Louisiana	240,000	129,000
Missouri	27,000	15,000
Mississippi	134,000	54,000
Tennessee	16,000	9,000
Total	637,000	331,000

^a Shrub/scrub habitats can include weedy agricultural fields, brushy swamps, overgrown watercourses, and regenerating forests.

^b Grassland habitats includes, pastures, hayfields, municipal grasslands (airstrips, golf courses, etc.), and levee berms.

Objective 2: Additional information is needed on the diversity and abundance of breeding birds using forest patches of different sizes and different forest types in the Mississippi Alluvial Valley. This could lead to development of models that predict bird distribution and abundance based on landscape and vegetative characteristics of forest patches.

Inventories of breeding and wintering birds are still needed for many managed areas. Established Breeding Bird Surveys (Robbins et al. 1986) should be conducted annually and completion of Breeding Bird Atlases encouraged. Standard avian monitoring techniques have been developed (Ralph et al. 1993) and point count methods refined for use in the Mississippi Alluvial Valley (Smith et al. 1993, Hamel et al. 1995). Breeding Bird Point Counts should be established and conducted annually in each of 50 disjunct stands with the following habitats: (1) semi-permanent flooded forests, such as those dominated by cypress or tupelo; (2) frequently flooded forests such as those dominated by overcup oak, water hickory, willow, and red maple; (3) infrequently flooded bottomland hardwood forests such as those dominated by red oaks, sweetgum, sugarberry, and ash; (4) upland forests located within or adjacent to the Mississippi Alluvial Valley [including pine, hardwood, and mixed pine-hardwood forests]; (5) early successional forests resulting from timber harvest or forest plantings; (6) grasslands; and (7) agricultural habitats. Until a regional or national repository for point count data is established, these data should be submitted in standard ASCII format (Hamel et al. 1995) to the Breeding Bird Point Count Repository, c/o Mark S. Woodrey, Mississippi Museum of Natural Science, 111 North Jefferson Street, Jackson, MS 39202.

Objective 3: Breeding bird densities in bottomland hardwood forests should be determined in order to assess the adequacy of forest patches to meet the area requirements of 500 breeding pairs of priority species. The assumption that the breeding densities reported in Hamel (1992b) are reflective of densities in the Mississippi Alluvial Valley requires validation. Research is needed to determine breeding bird densities (with associated confidence intervals) that reflect variability in patch characteristics and habitat types. Habitat within and among forest patches is clearly not uniformly suitable for breeding by a particular species and breeding bird densities are not constant over time and space.

Densities of birds can be assessed indirectly by using point count surveys in that accurate assessment of actual breeding bird density at a landscape scale (i.e., thousands of ha) is not economically feasible. The relative abundance determined from these surveys can be used as an index to the actual densities. Actual densities on small plots, such as 10 to 50 ha Breeding Bird Census (Hall 1964) plots used in demographic assessments, can be indexed to relative abundance estimates derived from point count surveys on those same sites (Hamel 1984). This tool can be used to extrapolate results of point count surveys over larger areas and among forest patches with different characteristics using ratio or regression estimators (Cochran 1977).

Objective 4: Quantification and modeling of the species-specific demographies of high priority forest bird populations is necessary in order the address the assumption that the proposed forest patches of 4,000, 8,000, and 40,000 ha will support source populations of targeted species. Species-specific demographic analyses conducted over a broad geographic range and over extended time periods will require information on nest survivorship, nest parasitism rates, nest predation rates, and re-nesting effort. Additionally, data are needed on age-specific survival of individuals, their dispersal, and philopatry. All these data are time consuming, difficult, and expensive to obtain, but they are essential to understanding the ability of forest patches to support populations of breeding birds.

Research designs that identify habitat use and dispersal of post-breeding adults and of young-of-the-year will be extremely useful in evaluating landscape conditions that promote persistence of source populations. Current assumptions about habitat use, patch characteristics, and survivorship are based upon knowledge of breeding adults. Few, or no, data exist on differences in habitat utilization between breeding adults and non-breeding adults during the breeding season, or between post-breeding adults and their offspring.

Logistic and fiscal constraints mandate collection of demographic data at only a relatively few locations. We assume, however, that demographic parameters related to forest patch characteristics can be extrapolated to other, similar forest patches within the Mississippi Alluvial Valley. To do this, however, we must be able to relate demographic parameters to forest patch metrics and within-patch vegetation characteristics.

Demographic studies are currently underway in bottomland hardwood forests in the Mississippi Alluvial Valley from southern Illinois to southern Louisiana. Current studies target different species, are conducted by different personnel, and have different overall objectives. However, these studies generally follow standardized methods of data collection to obtain estimates of productivity, nest survival, nest parasitism rates, and nest predation rates. Some of these data are from certain high priority species, such as the Cerulean Warbler, but the majority are from more common species such as Acadian Flycatchers and Prothonotary Warblers. Results of these studies may be applicable to other sites and to other species within these sites.

One of the reasons that survival of individuals, including age-specific estimates within and among seasons, is difficult to assess is a lack of data on the dispersal of individuals. Data on site fidelity and dispersal are similarly difficult to obtain and require extensive investigations across both area and time to estimate. Also important but even more difficult to measure is the impact of forest management on rates and distances of dispersal. Minimal estimates of survival of individuals and site fidelity, however, are being obtained through color-marking of individuals in mark-recapture studies or through constant-effort mist netting. Application of open population models (e.g., Pollock et al. 1990, Lebreton et al. 1992) to these data should result in estimates of survival with defined confidence limits.

Finally, predictive models need to be created that estimate species-specific demographic parameters based on the characteristics of the forest patch and under different forest management regimes. It may thereafter be possible to evaluate the assumption that demographic parameters can be predicted from forest patch metrics.

Objective 5: There should be tests of the assumption that the number of source populations proposed within the Mississippi Alluvial Valley constitutes an adequate number of breeding individuals within an overall meta-population to ensure long-term maintenance of genetic diversity and population viability. We are not now in position to test whether the proposed minimum of 500 breeding pairs is sufficient to ensure viability within a forest patch nor do we have sufficient data to support meta-population dispersal within the Mississippi Alluvial Valley. Data required to begin a meta-population analysis within the Mississippi Alluvial Valley include species-specific estimates of survival rates of hatching-year birds, rates and distances of dispersal of young from natal to breeding sites, among year dispersal of adult breeders, and rates of gene flow among breeding populations.

Other Objectives: Other research needs related to migratory landbirds include investigation of habitat use during migration, use of non-forested habitats by forest breeding birds, use and availability of food resources, and winter habitat use.

Specific research objectives, including investigators, study plans, cooperators, and time tables, have been identified for wintering waterfowl (Loesch et al. 1994). Similarly detailed research objectives are needed for shorebirds. In all cases, however, completion of proposed research and implementation of monitoring plans will require increased resource levels and as much, or more, cooperation as is currently the hallmark of this physiographic area.

Forest Openings, Edges, Early-Successional Shrub-Scrub

Status and importance

Within mature forests, small forest openings occur naturally in gaps created by fallen trees. Larger openings containing shrub-scrub habitat are created and maintained in these forests by disturbance phenomena such as: grazing, wind, tornados, hurricanes, ice storms, flooding, and most notably fire. Additionally, Yaich (in press) speculated that, formerly, passenger pigeon roosts, covering up to several thousand acres, caused catastrophic disturbances through breaking of tree limbs and guano deposition. Naturally occurring edge habitats are still associated with these disturbances but edge and early-successional habitats are now largely the result of human intervention. These disturbances all promote subsequent successional shrub-scrub development.

Earlier this century, land clearing for small farms and inefficient farming practices on these farms was common in the Mississippi Alluvial Valley. These small logging and farming operations provided an anthropogenic supplement to naturally occurring shrub-scrub habitats. Since the 1940's, small-scale timber harvests have been replaced by large-scale land clearing with subsequent conversion to large-scale agricultural operations. With more efficient farming techniques on these large cleared fields, much of the "old-field" and "hedgerow" habitats associated with small farms has been lost. Additionally, suburban development has encroached on cleared land, and forest succession has proceeded towards more mature forest stages. Currently, lands managed through even-aged silvicultural practices that remove all or most of the standing timber (i.e., "clearcut"), but are not converted to land uses other than forest production, provide most of the early successional shrub-scrub habitat within the Mississippi Alluvial Valley. As with natural disturbances, clearcuts provide transitory habitats and do not provide long-term stability for shrub-scrub species in any one tract. The current trend, however, is away from large clearcuts on both public land and non-industrial private lands. Similarly, the trend continues away from inefficient farming. Efforts to restore natural disturbance regimes to this bottomland ecosystem are minimal. Thus, active management to create and maintain early successional shrub-scrub habitats may be required to provide for those biotic communities and birds dependent on shrub-scrub habitats.

Although estimating the area of this relatively ephemeral habitat is difficult, classified TM imagery from 1992 suggests that >600,000 ha of scrub-shrub habitat are currently within the Mississippi Alluvial Valley (Table 9). This area includes weedy fields, idled farmland, overgrown drainages, and shrubby swampland as well as regenerating forests, but they do not include forest edge or thinned forest habitats that may account for an additional >300,000 ha in the Mississippi Alluvial Valley (see Table 8).

Table 9. Area of upland forest (in hectares) on ridges within the Mississippi Alluvial Valley classified from 1992 satellite imagery (USGS, unpublished data).
Physiographic Area (State)	Total Area	Forested Area
Crowley's Ridge Arkansas Missouri	145,000 55,000	84,000 17,000
Macon's Ridge Arkansas Louisiana	7,000 276,000	1,000 40,000
Sicily Island Louisiana	5,000	4,000
Total	488,000	146,000

Priority species, species suites, and habitat requirements

Several forest edge species, including Mississippi Kite, Orchard Oriole, and White-eyed Vireo, warrant management attention within the Mississippi Alluvial Valley. These species can also be found in the interior of bottomland forests, particularly where disturbance has opened the existing canopy. Forest edge species may be highly susceptible to nest depredation and parasitism as these habitats also harbor foraging cowbirds and nest predators (Robinson et al. 1995). Other species with high concern scores, such as Painted Bunting and Loggerhead Shrike, are more typically associated with early successional habitats or grass-dominated forest edges. Although limited in abundance and restricted in distribution to the far northern portions of the Mississippi Alluvial Valley, Bell's Vireo has a high concern score and may be of local concern in scrub-shrub habitats.

Population trends for widespread breeding species associated with shrub-scrub habitats exhibit an overall decline in the southeastern U.S., some of which are very steep. In addition to nongame species, range-wide declines in American Woodcock and Northern Bobwhite are

alarming. The habitat requirements of game bird populations should be addressed in conjunction with plans to formalize area, distribution, and rotation objectives for early-successional habitats.

Habitat and population objectives

Specific population goals have not been established but the tentative goal is to provide forest openings, edges, and early-successional habitats for scrub/shrub dependent species without subjecting forest-interior species in the Mississippi Alluvial Valley to increased nest depredation or parasitism. Because it takes little time to shift land use to a shrub-scrub condition, but conversion to mature forest takes decades, critical to considerations of area objectives for shrub/scrub species are the habitat objectives for species dependent upon more mature forests.

Within the southeastern United States, Capel et al. (1994) recommended a combination of early successional habitats to provide foraging, nesting, and cover needs of early-successional wildlife populations. Specifically, habitat objectives were set at 1 million ha of 5-year idled lands in native vegetation or grass-legume mixes, 1 million ha of annual vegetation (forbs or annually established cover), and 2 million ha of long-term (10-20 years) herbaceous or shrub cover. Long-term cover provides the greatest benefit to most shrub-scrub species, especially if managed with controlled burns rather than by mowing.

An appropriate allocation of the above recommendations has yet to be made within the Mississippi Alluvial Valley. However, based on current estimates of scrub-shrub habitat (>600,000 ha) and forest edge habitat (>300,000 ha), an overall habitat objective of 1 million ha may not be unreasonable, particularly if long term scrub/shrub habitats can be positioned to provide gradual (soft) edges to mature forest patches.

Implementation recommendations and opportunities

Management strategies to provide suitable forest edge and early-successional habitats should minimize detrimental impacts on birds dependent upon mature forests. Land ownership and silvicultural objectives may influence which management strategies are employed. Regeneration on large consolidated clearcuts (e.g., 10-20 acres on public lands, 50-100 acres on private lands) will provide suitable habitat for shrub-scrub dependent birds but, in small forest patches, openings of this magnitude may negatively impact forest-interior breeding birds. Conversely, a similar area of small gaps within a single forest patch may be equally detrimental to forest-interior species. Undoubtedly, the impact of area of harvest (i.e., gap size) and harvest frequency on the reproductive success of shrub-scrub species, as well as forest-interior species, requires additional research before specific recommendations can be made.

Over the next few decades, much early successional habitat will be created through reforestation associated with implementation of Wetland Reserve and Conservation Reserve Programs and through reforestation on public lands. In the long-term, however, we anticipate most shrub-scrub habitat will result from harvested and subsequently regenerated forests.

Towards this end, opportunities to create large blocks of early-successional habitat on private lands exist through agro-forestry (e.g., short-rotation, planted or coppiced cottonwoods or other fast-growing hardwoods such as sycamore [Twedt and Portwood, in press]). Large blocks of early-successional habitat can be provided in these forest-farms on a continuing basis as trees are harvested and subsequently regenerated. Currently, most short-rotation forest production goes into pulpwood but the potential exists for increased area devoted to woody crop production intended for biomass production (Ranney et al. 1985). Agro-forestry should be promoted as a replacement for annual agricultural crops, such as cotton and soybean, but not as a replacement for existing or potential forests managed for saw-timber. When possible, forest farms should be located adjacent to existing mature forests, under longer rotation silvicultural management, to provide a buffer against nest predation and parasitism within the more mature forest stands, thereby increasing the effective "interior area" of the forest patches.

When early-successional habitat objectives targeting a continuing flow of early-successional habitats within the Mississippi Alluvial Valley are finalized, these objectives should be allocated to individual states and, where appropriate, to individual bird conservation areas.

Evaluation of assumptions

Species densities and demographic data are needed for species breeding at the interface of forest and agriculture ("hard" forest-edges), for species breeding in early-successional habitats, and for species breeding at the interface of mature forests and early-successional habitats ("soft" forest-edges). Of particular relevance is an examination of the ability of shrub/scrub buffers to mitigate negative impacts of predation and parasitism on interior nesting species. These parameters should be investigated in a variety of landscape contexts, particularly in largely forested and largely non-forested landscapes, to develop sound recommendations that can be adapted to the objectives of different landowners. Short-rotation managed forests may provide opportunities for manipulative experimentation to address issues such as population demographics, dispersal, and provide insight into the fates of birds displaced by timber harvest..

Upland Oak-Hickory Hardwoods

Status and importance

The most striking upland forests within the Mississippi Alluvial Valley are located on ridges and domes embedded within this floodplain (e.g., Crowley's Ridge, Macon's Ridge, and Sicily Island). Other oak-hickory hardwood forests (Bryant et al. 1993, Skeen et al. 1993) are found along the edges of the valley in loess bluffs (this habitat is covered within the East Gulf Coastal Plain conservation plan) and otherwise localized in small stands along the highest ridges within forests dominated by bottomland tree species. Upland woodlands on braided stream terraces may be the most poorly represented habitat on public lands in the Mississippi Alluvial Valley (M. Swan, pers. com.).

Crowley's Ridge is a high ridge completely embedded within the Mississippi valley that runs north to south from Missouri through Arkansas. Macon's Ridge, running from southeast Arkansas southwesterly through northeast Louisiana, is wider and lower than Crowley's Ridge. Sicily Island is a dome cut off from the Valley's western bluff by the Quachita River floodplain. Because these ridges are on high, often sloping ground, they are generally unsuitable for large-scale agriculture but are instead heavily spotted with homes, pastures, small fields, and even pine stands, leaving the remaining hardwoods in a highly fragmented state. Of the 487,655 ha of uplands comprising Sicily Island, Crowley's Ridge, and Macon's Ridges, 146,405 ha remain forested (Table 10). Nevertheless, neotropical migratory landbird species that are not expected within most of valley's forested wetlands occur where relatively large, mature hardwood forest patches remain on these ridges.

The upland forests on the eastern loess bluff of the Mississippi Alluvial Valley remain relatively intact and predominately in hardwood forests. Because underlying soils are highly erodible, large scale deforestation of the loess bluff forests is not likely. However, potential threats to bluff hardwood forests include increasing urbanization, fragmentation from small agricultural fields and pastures, and conversion to managed pine forests.

Table 10. Twenty-five years (1966-1991) of Breeding Bird Survey population trends for the southeastern U.S. among widespread grassland birds found in the Mississippi Alluvial Valley.
Species	Population trend (% / yr)	Number of Routes	Proportion of Decreasing Routes	Mean Birds per Route
Killdeer Charadrius vociferus	0.1	807	0.48	4
Eastern Kingbird Tyrannus tyrannus	-1.0	784	0.57^**	5
Horned Lark Eremophila alpestris	-1.5^*	398