Advertisement

Hydrobiologia

, Volume 804, Issue 1, pp 119–137 | Cite as

Waterbird response indicates floodplain wetland restoration

  • Heath M. Hagy
  • Christopher S. Hine
  • Michelle M. Horath
  • Aaron P. Yetter
  • Randolph V. Smith
  • Joshua D. Stafford
RIVER FLOODPLAIN RESTORATION

Abstract

Despite extensive anthropogenic degradation of most wetlands and other aquatic habitats associated with large rivers in the Midwest, the region still supports continentally important numbers of waterbirds during autumn and spring migration; however, few data exist to evaluate wetland restoration success and identify thresholds where changes in management may be necessary to meet conservation targets. We tracked waterbird response to restoration of a historical floodplain wetland complex along the Illinois River during 2007–2013 relative to waterbird use of other wetlands and floodplain lakes in the region. Dabbling ducks and other waterbirds showed dramatic responses to restoration, each accumulating more than 3 million use-days/year and comprising more than 30% of the total waterbird use-days in the Illinois River Valley during autumn and spring migrations. We identified use that was strongly disproportionate to availability within the region for several waterbird taxa and documented nesting by several species of conservation concern. Many species and foraging guilds of waterbirds [e.g., American coot (Fulica americana), dabbling ducks (Anatini)] responded rapidly to wetland restoration, continued to use Emiquon Preserve regardless of changing conditions at reference sites, and showed relatively limited temporal variation, thereby demonstrating their utility as indicators of habitat conditions and restoration trajectory.

Keywords

Emiquon Preserve Restoration Waterfowl Waterbirds Illinois river 

Introduction

The Upper Midwest region of the United States has experienced extensive wetland loss, with most states losing more than 70% of their original wetland area during the last two centuries (Dahl, 1990). Most large rivers in the region have been isolated from portions of their floodplains by levees and associated wetlands and backwater lakes drained for agriculture (Tockner & Stanford, 2002), filled with sediments (Bellrose et al., 1983), colonized by invasive species (Bajer et al., 2009), or degraded in other ways (Jackson & Pringle, 2010). In particular, the Illinois River Valley (IRV) was home to a diverse array of wetlands filled with aquatic vegetation prior to the twentieth century (Bellrose et al., 1983). By the 1920s, >50% of the Illinois River floodplain from LaSalle to Grafton, IL had been isolated behind levees and aquatic areas drained, and by the 1950s, virtually all of the obligate aquatic vegetation within portions of the floodplain connected to the river had been lost (Mills et al., 1966; Bellrose et al., 1983; Sparks, 1984).

Although the IRV remains a continentally important migration stopover for migratory birds, abundances of waterbirds have declined since the 1950s (Havera, 1999; NAWMP, 2012; Smith et al., 2012). For example, more than 1.6 million mallards (~20% of the long-term average breeding population in North America; Anas platyrhynchos) were counted during aerial inventories in the IRV in 1948, and peak numbers of lesser scaup (Aythya affinis) exceeded 500,000 prior to the mid-1950s (Havera, 1999; USFWS, 2014). During 2010–2014, aerial surveys conducted by the Illinois Natural History Survey detected a mean peak abundance of approximately 246,000 mallards and 13,000 lesser scaup, illustrating the regional decline of these two species. However, other species of dabbling ducks have increased in recent years, such as northern pintail (Anas acuta; 40% increase since the 1950s) and American green-winged teal (A. carolinensis; 800% increase since the 1950s), concurrent with significant wetland restoration and enhancement efforts in the IRV (INHS, unpublished data).

The Upper Mississippi River and Great Lakes Region Joint Venture created by the North American Waterfowl Management Plan considers the IRV a focal region to provide habitat for millions of waterfowl during spring and autumn migrations (Soulliere et al., 2007; NAWMP, 2012). During autumn, dabbling ducks stop in the IRV for approximately 28 days (O’Neal et al., 2012), and some species may remain longer than 2 months before migrating southward (Hagy et al., 2014). Recent research has demonstrated that migration stopover duration is related positively to foraging habitat quality, which incentivizes wetland managers to create foraging habitats for waterbirds during autumn and spring (O’Neal et al., 2012). Most aquatic vegetation communities, especially submersed and emergent macrophytes, have been eliminated from wetlands in the IRV, the pools of the middle Mississippi River, and many other areas of the Upper Midwest important to waterbirds during migration (Moore et al., 2010; Stafford et al., 2010). Wetland restoration goals include restoring and enhancing approximately 120,000 ha of shallow semi-permanent marsh for waterfowl in the Upper Midwest to increase foraging habitat availability (Soulliere et al., 2007), but few indicators have been verified to characterize success of wetland restoration projects in providing high-quality habitat for focal species.

Wetland use by waterbirds serves as an indicator of wetland condition, succession, hydrologic modifications, and restoration success (Austin et al., 2001; Gawlik, 2006; Kajtoch et al., 2014; Figarski & Kajtoch, 2015). During autumn migration, sanctuary conditions (Hagy et al., 2016), vegetation community composition and interspersion (Stafford et al., 2007, 2010), foraging habitat quality (O’Neal et al., 2012), and landscape composition (Beatty et al., 2014) tend to be primary predictors of wetland use and abundance by waterbirds in the Upper Midwest. Great diversity in food habits (Callicutt et al., 2011), habitat selection tendencies (Webb et al., 2010), tolerance to disturbance (Blumstein et al., 2005; Hagy et al., 2016), migration phenology (Havera, 1999), and other life-history characteristics among taxa of migrating waterbirds in North America allow responses of different taxa to wetland restoration to be used as indicators of habitat structure (Paracuellos, 2006), vegetation structure (Weller & Fredrickson, 1974; Kaminski & Prince, 1981; Linz et al., 1996), or colonization by invasive species and overall wetland quality (Bajer et al., 2009). Waterbirds are likely suitable indicators of specific biological characteristics, ecological processes, and wetland succession stages that can be used to evaluate wetland restoration and management progress along with a diverse suite of other metrics (Parrish et al., 2003). Although a number of conservation programs and large initiatives have restored portions of floodplains along large rivers, few data exist that can be used to identify acceptable ranges in waterbird use, vegetation communities, and water quality where changes in management may be necessary to meet conservation targets (King et al., 2006; Stafford et al., 2007; Bajer et al., 2009). Identification of thresholds in values of indicators is important so that adaptive management can mitigate negative effects of undesirable shifts in communities or ecological states (Suding et al., 2004; Tierney et al., 2009a). Moreover, results of restoration projects along with evaluation metrics and acceptable ranges of indicators should be reported in the scientific literature to allow comparisons to other restoration projects and reference areas, avoid perpetuation of restoration practices that do not yield acceptable outcomes, and advance the field of restoration ecology (Lake, 2001).

We evaluated the response of waterbirds to floodplain wetland restoration at Emiquon Preserve during 2007–2013 to track restoration success relative to target conditions established by The Nature Conservancy (c.f. Hine et al., 2016; Van Middlesworth et al., 2016). Waterbirds and their foraging habitats were identified as potential indicators of restoration success and habitat trajectory under the relevant key ecological attributes which were established prior to the initiation of restoration at Emiquon Preserve (Parrish et al., 2003; TNC, 2006). Comparative data on waterbird use of other lakes and wetlands across a range of conditions in the region were available from the Illinois Natural History Survey, which made this indicator especially important and appropriate for use. We evaluated (1) abundance, diversity, and behavior of waterfowl and other waterbirds during spring and autumn migrations and (2) productivity using brood counts as an index during summer. We hypothesized that waterbirds would quickly colonize wetland habitats and taxonomic composition would change over time in response to changes in habitat structure, especially vegetation communities and other cover types (Hine et al., 2016). We hypothesized that waterbird productivity would increase over time as waterbird species colonized emergent vegetation communities and other habitats. We predicted that diving and dabbling duck use would be two consistent indicators of restoration progress relative to other degraded and previously restored aquatic habitats in the IRV.

Methods

Study area

Emiquon Preserve is a 2,723-ha floodplain and associated upland restoration adjacent to the Illinois River owned and managed by The Nature Conservancy (Fig. 1). The historical floodplain area encompassed within Emiquon Preserve (~2,400 ha) was disconnected from the Illinois River in the early 1920’s and several historical backwater lakes and associated wetlands were drained and subsequently farmed until 2006, when the restoration began. Pumping of tile drainage ditches into the Illinois River was ceased in spring 2007 and water began accumulating in the historical floodplain lake basins behind the drainage and levee district levee (Havera et al., 2003). Limited planting of native vegetation was conducted in a few areas of Emiquon Preserve, and most vegetation appears to have grown naturally (Hine et al., 2016). During 2007–2013, Emiquon Preserve primarily comprised submersed and floating-leaved aquatic vegetation communities (mean 44%), open water (mean 20%), persistent emergent vegetation communities within dense (mean 10%) and hemi-marsh (mean 10%) configurations, and non-persistent emergent vegetation communities (mean 9%) and remained disconnected from the Illinois River by levees (Hine et al., 2016). Maximum water depth within Emiquon Preserve during our study period was approximately 4 m and hydrologic variation was small relative to other nearby wetlands and lakes connected to the Illinois River (Sparks et al., 2016).
Fig.. 1

Aerial survey locations (white circles) for waterbirds along the LaGrange and Peoria pools of the Illinois river and the location of Emiquon Preserve in Fulton County, Illinois, USA with general locations of brood survey observation points (black circles)

Study design

During autumns 2007–2009 and springs 2008–2013, we enumerated waterbirds by species with a spotting scope or binoculars from fixed, elevated vantage points to reduce potential visibility bias from emergent vegetation. We initiated autumn surveys in early September, as waterfowl began migrating through the region, and terminated counts following freeze-up, which typically occurred in December after waterbird use largely ceased. Spring surveys began when ice receded (i.e., mid-February or early March) and concluded following spring migration (i.e., mid-April) each year. We conducted ground surveys of waterbirds approximately weekly. We also enumerated waterbirds by species during autumns 2007–2013 in conjunction with the Illinois Natural History Survey’s (INHS) autumn waterfowl aerial inventories (Havera, 1999). Aerial inventories of 23 lakes and wetland complexes along and nearby the Illinois River that comprise most waterfowl use in the region were conducted approximately weekly using a fixed-wing, single-engine aircraft at altitudes of 60–140 m and speeds of 160–240 km/h using a single observer (Stafford et al., 2007; Fig. 1).

We converted weekly raw counts for each survey site to use-days to evaluate overall waterbird use of Emiquon Preserve each year (Stafford et al., 2007). Use-days are estimates of abundances extrapolated over a period of interest (i.e., fall or spring) and are typically used to estimate carrying capacity or compare use of different wetlands over time (Soulliere et al., 2007). For example, 100 birds using a wetland for 10 days equates to 1,000 use-days. We used concurrent aerial inventory data from 23 backwater lakes and wetland complexes located along the Illinois River, which accounted for approximately 90% of IRV peak duck abundances, to compare to use-days among survey locations within the region (Havera, 1999; Horath & Havera, 2002). Count data from Emiquon Preserve was seperated from nearby wetlands resulting in a total of 24 survey locations. Lastly, we expressed duck use estimates as use-days per hectare of wetland (UDs/ha) to standardize for changing wetland area at Emiquon Preserve across years and different sizes of inventory locations within the region. We modeled use-day densities by year and foraging guild [i.e., dabbling ducks (Anatini), diving ducks, geese, and waterbirds other than waterfowl (other waterbirds)] using linear mixed models with year as a repeated measure. Additionally, we calculated Simpson’s diversity for each INHS aerial inventory location within the IRV for comparison with Emiquon Preserve during our study period. We used linear regression to compare Simpson’s diversity and the ranking of this measure in separate models with year (Johnson, 1980).

To address our hypothesis that waterbird use would be a suitable indicator of restoration progress and relative habitat quality, we compared use-days at Emiquon Preserve with Chautauqua National Wildlife Refuge (CNWR) and other lakes and wetlands in the IRV. The south pool of CNWR is primarily managed for moist-soil vegetation and was restored by the United States Fish and Wildlife Service and the Army Corps of Engineers through the Habitat Rehabilitation and Enhancement Program during the late 1990s (Bowyer et al., 2005). This wetland typically provides high-quality habitat for migratory birds through sanctuary and abundant forage (Hagy et al., 2012, 2016) and is adjacent to Emiquon Preserve providing a geographically unbiased reference site. We compared use-days by bird guild using two-tailed t-tests for means between Emiquon Preserve and nearby CNWR. Furthermore, we assigned cover types to aerial survey sites of the INHS throughout the IRV using aerial photographs, foraging habitat quality scores (e.g., O’Neal et al., 2012), or other available information according to dominant vegetation community or other cover type (i.e., dense emergent marsh associated with aquatic bed vegetation, moist-soil and non-persistent emergent vegetation, or open water associated with mud flats; Hine et al., 2016) where birds typically occurred. We modeled total waterbird use-day density as a function of dominant cover type with year as a repeated measure and site as a random effect using linear mixed models.

Due to variable distances between observation points and the wetland edge and the gregarious nature of waterfowl, we were unable to use traditional distance sampling to estimate detection probability during ground surveys. Instead, we compared concurrent weekly aerial and ground counts by species from autumns 2007–2009 and assumed that aerial counts approximated complete counts relative to detection probability because limited visibility bias due to emergent vegetation should have occurred (Stafford et al., 2007). We used a paired, two-tailed t test to determine if aerial and ground counts were similar.

Concurrent with spring ground surveys, we conducted behavioral observations using instantaneous scan sampling to evaluate the functional response of ducks over time and across different habitats (Altmann, 1974; Paulus, 1988). We collected 5–10 scan samples during each ground count on species that were present at the wetland throughout the migration period. During each scan sample, we collected species, sex, and an instantaneous behavior on at least 50 individuals of the same species [i.e., feeding, resting, social (e.g., courtship and aggression), locomotion (e.g., swimming, walking, and flying), and other (e.g., comfort and preening)]. As instantaneous scan sampling can underestimate foraging time for diving ducks (Hohman, 1984; Baldassarre et al., 1988), we watched each diving duck for approximately 10 s and recorded “foraging” if the duck dove during that period and recorded the original behavior observed if the duck did not dive during that period. Because foraging was the most common behavior and of most interest relative to foraging habitat quality, we used a linear mixed model to determine if foraging effort differed among cover types (Hine et al., 2016), sex, or foraging guild and designated year as a random effect and week as the repeated measure.

During late spring and summer 2008–2013, we conducted bi-weekly brood surveys from 4 elevated observation points surrounding Emiquon Preserve which served as an index of waterbird productivity. All fixed-point surveys began at sunrise and lasted for 1 h to coincide with a period of increased brood activity (Ringelman & Flake, 1980; Rumble & Flake, 1982). During each survey, we continually scanned wetland habitat using spotting scopes and binoculars and documented species, number of young and adults, and brood age class of all waterbirds (Gollop & Marshall, 1954). Because broods may have been difficult to detect within emergent vegetation, we used active flush counts twice during 2008, immediately following fixed-point surveys, to determine if passive surveys could be used as an index for total brood abundance (Rumble & Flake, 1982). During flush surveys, 2 observers traversed along opposite lake margins and flushed broods from emergent vegetation to open water where they were enumerated by species similar to fixed-point surveys. We used paired t-tests to determine if the number of broods observed differed between methods for each species observed and overall. We converted brood counts to densities by dividing the number of broods observed by the survey area. To account for detection probability, we used published detection rates from mallards to adjust count data (Pagano & Arnold, 2009). Subsequently, we used separate simple linear regression models to evaluate changes in brood density across years, if mean age of broods changed within years, if the number of young per brood changed within years, and if the number of young per brood changed across years.

Prior to all analyses, we examined data histograms, variances of independent variables, and plots of residuals to ensure data met assumptions of analyses and used recommended data transformations as needed (Zar, 2009). When using mixed models with repeated measures, we used Akaike’s Information Criterion to select an appropriate covariance structure and specified restricted maximum likelihood estimation of fixed effects (Littell et al., 2006). We determined α = 0.05 a priori, performed Tukey’s pair-wise multiple comparisons tests of means among levels of categorical variables when P ≤ 0.05, and calculated means with standard errors from untransformed data.

We identified potential acceptable ranges for indicators based on 50th and 80th percentiles of each indicator across our study years and INHS aerial survey locations. Based on use-day densities from Emiquon Preserve and INHS aerial surveys elsewhere in the region, we assigned rankings and use-days ranges to each indicator. A ranking 1–5 in raw use-days indicated selection of Emiquon Preserve relative to other INHS autumn survey locations in the IRV, based on average wetland area during 2008–2013. A ranking of 6–12 indicated Emiquon Preserve was approximately within the top 50% of other surveys sites, whereas a ranking > 12 signified use at Emiquon Preserve was less than 50% of survey sites in the IRV during autumn. Accompanying use-day ranges were based on the approximate 50th (12th ranking of 24 sites) and 80th (5th ranking of 24 sites) percentiles of use-day densities in the IRV during autumns 2007–2013. Concurrent aerial survey data from other aquatic habitats in the region were only available for diving ducks during spring; thus, we scaled rankings according to the number of sites surveyed during 2012–2014 for diving ducks and present ranges based on the 50th and 80th percentiles of use at Emiquon during springs 2008–2013. Foraging rates were similarly scaled from percentile data collected at Emiquon Preserve during our study period. We assumed that foraging rate was related positively to habitat quality (Lyons, 2005).

Results

Waterbird use

Autumn

We conducted 25 ground surveys during autumns 2007–2009 within 3 days of INHS aerial surveys to determine if methodological differences influenced total waterbird counts. Ground surveys produced similar counts as aerial surveys (t = 1.71, P = 0.147). Subsequently, we report use-days calculated from aerial inventories during autumns 2007–2013.

The most abundant species by total use-days during autumn were American coot (Fulica americana; mean 45.3%), mallard (mean 9.8%), northern pintail (mean 9.3%), American green-winged teal (mean 8.8%), gadwall (Anas strepera; mean 8.2%), northern shoveler (Anas clypeata; mean 5.4%), blue-winged teal (Anas discors; mean 5.0%), ruddy duck (Oxyura jamaicensis; mean 2.5%), ring-necked duck (Aythya collaris; mean 1.9%), and American wigeon (Anas americana; mean 1.6%)—all other species represented less than 1% of total use-days. For these common species, only gadwall (F = 12.3, P = 0.017), northern shoveler (F = 6.8, P = 0.047), and canvasback (F = 23.7, P = 0.004) increased their use of Emiquon Preserve over time—use by other species remained similar across years (P > 0.091). Peak counts of all waterbirds and waterbirds other than dabbling ducks occurred on a mean date of 27 October, whereas peak counts of geese (mean 11 days) and other waterbirds (mean 1 day) occurred earlier, and dabbling ducks other than mallards (mean 1 day), all dabbling ducks (mean 7 days), and diving ducks (mean 7 days) occurred later (Table 1).
Table 1

Mean use-days (UD; ±SE), percent of total use-days observed across waterbird species, densities (UDs/ha ± SE), and results from a two-tailed t test comparing autumn and spring UDs of waterbirds across years at Emiquon Preserve in Fulton County, Illinois, USA during autumn and spring migrations, 2007–2013

Species

Autumn

Spring

P

UD

SE

%

UD/ha

SE

UD

SE

%

UD/ha

SE

American coot

2,697,064

478,682

45.3

1,792.0

144.0

605,044

166,297

24.2

348.2

88.5

<0.001

Mallard

582,442

78,111

9.8

495.6

128.3

152,412

26,149

6.1

85.8

12.0

0.014

Northern pintail

552,951

103,479

9.3

447.6

113.6

39,311

14,784

1.6

22.4

7.4

0.006

American green-winged teal

525,040

65,733

8.8

471.7

160.7

84,306

32,649

3.4

58.9

30.5

0.040

Gadwall

486,509

74,606

8.2

351.2

48.6

115,306

35,445

4.6

66.4

19.3

<0.001

Northern shoveler

320,479

55,186

5.4

254.0

64.6

193,763

34,766

7.8

108.2

16.3

0.067

Blue-winged teal

297,227

75,350

5.0

229.4

65.2

28,807

9,494

1.2

17.2

5.6

0.012

Rudy duck

147,248

48,700

2.5

84.1

24.4

186,197

20,554

7.5

105.6

8.3

0.452

Ring-necked duck

113,770

46,713

1.9

61.5

24.6

152,582

48,045

6.1

98.8

38.1

0.416

American wigeon

95,662

19,929

1.6

65.0

8.5

16,643

6,058

0.7

11.6

5.9

<0.001

American white pelican

34,769

11,542

0.6

18.5

6.2

12,393

4,838

0.5

6.6

2.4

0.124

Canvasback

24,151

7,096

0.4

12.7

3.7

29,364

7,976

1.2

19.5

7.7

0.420

Canada goose

16,544

4,951

0.3

10.9

2.5

15,068

5,020

0.6

8.4

2.4

0.501

Double-crested cormorant

13,033

3,835

0.2

6.9

2.0

14,109

5,799

0.6

7.5

3.0

0.875

Lesser scaup

11,014

5,257

0.2

6.0

2.8

181,885

46,551

7.3

115.9

38.9

0.011

Bufflehead

8,271

3,116

0.1

4.4

1.7

32,718

5,376

1.3

19.5

3.7

0.002

Common merganser

492

238

0.0

0.3

0.1

23,420

12,539

0.9

12.7

6.8

0.071

Red-breasted merganser

     

1,563

628

0.1

0.8

0.3

Hooded merganser

4,838

3,605

0.1

2.6

2.0

4,016

1,457

0.2

2.2

0.8

0.845

Lesser snow goose

4,089

2,642

0.1

2.2

1.5

529,708

173,316

21.2

282.9

93.9

0.008

Common goldeneye

3,343

1,046

0.1

2.0

0.6

14,615

4,133

0.6

8.1

2.2

0.014

Greater white-fronted goose

3,290

1,226

0.1

1.8

0.7

40,736

15,865

1.6

21.7

8.3

0.024

American black duck

2,177

1,088

0.0

1.2

0.5

67

31

0.0

0.0

0.0

0.063

Redhead

2,086

1,399

0.0

1.1

0.7

12,277

6,278

0.5

8.8

5.1

0.138

Tundra swan

1,110

528

0.0

0.6

0.3

Trumpeter swan

1,007

392

0.0

0.5

0.2

Mute swan

344

148

0.0

0.2

0.1

Wood duck

516

245

0.0

0.3

0.1

White-winged scoter

14

14

0.0

0.0

0.0

Other

0.1

2.9

1.9

6,279

5,458

0.3

5.4

5.0

Dabbling duck

2,862,487

321,658

48.1

2,315.7

541.8

631,130

73,675

25.3

370.8

50.3

0.007

Diving duck

309,884

97,220

5.2

171.8

49.8

594,673

84,062

23.8

364.3

79.1

0.057

Geese

23,923

8,205

0.4

14.9

4.1

585,511

183,992

23.5

313.0

99.1

0.007

Other dabbling duck

2,280,045

298,536

38.3

1,820.1

432.8

478,718

56,952

19.2

285.1

45.5

0.008

Other waterbird

2,744,866

487,257

46.1

1,817.6

141.6

631,546

165,706

25.3

362.3

87.5

<0.001

Total waterbird

5,952,714

822,144

100.0

4,326.2

594.8

2,495,579

320,044

100.0

1,444.7

176.9

0.001

Use-day density during autumns 2007–2013 varied by foraging guild (F = 49.1, P < 0.001) but not year (F = 0.4, P = 0.883; Fig. 2). Use by dabbling ducks (mean 2,316 ± 542 UD/ha) was greater than diving ducks (mean 172 ± 50 UD/ha; t = 5.9, P < 0.001) and geese (mean 15 ± 4 UD/ha; t = 10.34, P < 0.001) but similar to other waterbirds (mean 1,818 ± 142 UD/ha; t = 0.3, P = 0.770; Table 1). Use by other waterbirds was also greater than geese (t = 10.1, P < 0.001) and diving ducks (t = 5.6, P < 0.001). Diving duck use was greater than geese (t = 4.4, P < 0.001). Notably, American coots used Emiquon Preserve more than any other waterbird species during autumn migration, averaging 2,697,064 use-days, which equaled 71% of total use-days of this species in the IRV (mean 1,792 ± 125 UDs/ha). Emiquon Preserve also attracted other notable species and guilds, such as dabbling ducks other than mallards (mean 1,820 ± 433 UD/ha), double-crested cormorants (Phalacrocorax auritus; mean 7 ± 2 UDs/ha; 3rd highest density in the IRV), and American white pelicans (Pelecanus erythrorhynchos; mean 19 ± 6 UDs/ha; 9th highest density in the IRV).
Fig. 2

Duck use-day densities (UD/ha) for foraging guilds of waterbirds during autumn (dark gray) and spring migration (light gray) at Emiquon Preserve in Fulton County, Illinois, USA during 2007–2013

More than 30% of total waterbird use-days in the IRV were observed at Emiquon Preserve during autumns 2007–2013, despite accounting for only approximately 5% of the flooded area surveyed. Autumn dabbling duck (mean 21%), diving duck (mean 25%), other dabbling duck (mean 37%), and other waterbird (mean 63%) use-days were also greater than expected relative to proportional area surveyed (Tables 2, 3). Furthermore, Emiquon Preserve ranked the highest in total use-days in 6 of 7 years and highest in use-day density in 3 of 7 years and never was ranked lower than 8th for dabbling ducks of all surveyed locations in the IRV. Use-days at Emiquon Preserve were consistently among the highest in the IRV for northern pintail, American green-winged teal, American wigeon, gadwall, northern shoveler, American coot, other waterbirds, other dabbling ducks, and total waterbirds. Proportional use-days were generally greater than, and rankings generally less than, other IRV survey locations suggesting selection in most years by most taxa (Table 2, 3). On average, Emiquon Preserve ranked 8th in Simpson’s diversity among 24 sites inventoried during autumn by the INHS (mean 0.75), and neither diversity score (F = 0.53, P = 0.498) nor ranking (F = 4.13, P = 0.096) varied by year (Table 3).
Table 2

Rank (mean from 2007 to 2013) and the percent of total waterbird use-days at Emiquon Preserve in Fulton County, Illinois, USA of species comprising >1% of use relative to other locations aerially surveyed by the Illinois Natural History Survey during autumn migration and early winter (September–January) within the Illinois River Valley (n = 24)

Year

American green-winged teal

American coot

American wigeon

Blue-winged teal

Gadwall

Mallard

Northern pintail

Northern shoveler

Ring-necked duck

Ruddy duck

Availabilitya

Rank

 2007–2013

1

1

1

1

1

6

1

1

6

2

9

Proportion of use-days

 2007 (%)

33.0

50.1

40.9

79.1

21.0

4.8

26.8

91.9

0.0

4.3

1.0

 2008 (%)

37.8

93.0

59.3

54.4

44.5

12.3

45.7

85.4

28.9

48.8

3.8

 2009 (%)

46.7

84.7

60.9

58.4

47.8

10.0

37.0

74.5

28.3

68.6

6.3

 2010 (%)

40.0

84.1

64.0

68.6

46.8

11.7

39.5

59.4

25.8

38.6

6.6

 2011 (%)

23.9

68.6

75.7

53.4

41.7

12.4

36.6

44.6

29.8

30.7

6.3

 2012 (%)

24.2

62.5

80.7

81.2

19.2

5.3

21.1

34.2

3.7

23.0

6.2

 2013 (%)

7.4

50.7

35.3

29.4

19.1

4.1

17.9

24.6

14.6

8.3

6.7

 2007–2013 (%)

30.4

70.5

59.5

60.6

34.3

8.7

32.1

59.2

18.7

31.8

5.3

aRank relative to other IRV wetlands or proportion of the total flooded area at Emiquon preserve relative to overall flooded area surveyed within the IRV

Table 3

Rank (mean from 2007 to 2013) and the percent of total waterbird use-days and Simpson’s diversity index score and ranking at Emiquon Preserve in Fulton County, Illinois, USA of foraging guilds relative to other locations aerially surveyed by the Illinois Natural History Survey during autumn migration and early winter (September–January) within the Illinois River Valley (n = 24)

Year

Dabbling ducks

Other dabbling ducksa

Diving ducks

Geese

Other waterbirds

Total waterbirds

Availabilityb

Simpson’s Diversity

Rank

 2007–2013

1

1

3

7

1

1

9

8

Proportion of use-days

 2007

15.0%

33.2%

1.2%

1.4%

40.7%

17.1%

1.0%

0.82

 2008

22.5%

45.8%

36.2%

0.8%

82.6%

31.5%

3.8%

0.75

 2009

26.0%

51.3%

41.9%

6.5%

75.3%

43.1%

6.3%

0.67

 2010

32.6%

47.8%

29.3%

10.7%

71.2%

42.4%

6.6%

0.76

 2011

25.4%

35.3%

34.1%

10.5%

63.1%

34.0%

6.3%

0.78

 2012

15.8%

26.1%

14.7%

13.9%

59.8%

23.1%

6.2%

0.78

 2013

11.7%

16.6%

14.2%

3.5%

47.2%

19.3%

6.7%

0.70

 2007–2013

21.3%

36.6%

24.5%

6.7%

62.8%

30.1%

5.3%

0.75

aDabbling duck species other than mallard

bRank relative to other IRV wetlands or proportion of the total flooded area at Emiquon Preserve relative to overall flooded area surveyed within the IRV

During autumns 2007–2013, use-day density at Emiquon Preserve exceeded estimates from the previously restored and managed south pool of CNWR in most years for several species and foraging guilds, such as American coot (+94%; P < 0.001), canvasback (+95%; P = 0.007), American wigeon (+83%; P = 0.020), blue-winged teal (+96%; P = 0.005), diving ducks (+84%; P = 0.017), other waterbirds (+93%; P < 0.001), and total waterbirds (+71%; P = 0.006). However, use-day densities of mallard (P = 0.821), dabbling ducks (P = 0.169), non-mallard dabbling ducks (P = 0.076), northern pintail (P = 0.178), American green-winged teal (P = 0.221), and geese (P = 0.334) were similar between Emiquon Preserve and the south pool of CNWR across years. Mean densities across years were never significantly greater at CNWR than Emiquon Preserve. Across all sites surveyed for waterbirds during autumn in the IRV (n = 23), Emiquon Preserve and other sites with emergent marsh vegetation communities accumulated most use-days during 2007–2013 (mean 54.2% ± 4.6% mean use-day density; F = 24.9, P < 0.001), followed by moist-soil and other non-persistent emergent vegetation (mean 35.1 ± 2.6%) and open water and mudflat cover types (mean 15.7 ± 1.9%).

Spring

We conducted 50 ground inventories of waterbirds during springs 2008–2013. The most abundant species were American coot (mean 24.2%), lesser snow geese (Chen caerulescens; mean 21.2%), northern shoveler (mean 7.8%), ruddy duck (Oxyura jamaicensis; mean 7.5%), ring-necked duck (mean 6.1%), mallard (mean 6.1%), gadwall (mean 4.6%), American green-winged teal (mean 3.4%), greater white-fronted goose (Anser albifrons; mean 1.6%), northern pintail (mean 1.6%), bufflehead (Bucephala albeola; mean 1.3%), canvasback (mean 1.2%), and blue-winged teal (mean 1.2%)—all other species accounted for <1% of total use-days. Dabbling ducks (mean 25.3%), geese (mean 23.5%), and diving ducks (mean 23.8%) each comprised approximately one quarter of birds counted at Emiquon Preserve during spring. Peak counts typically occurred around 9 March, with peak counts of geese occurring earlier (mean 8 days) and dabbling ducks (mean 15 days), diving ducks (mean 11 days), dabbling ducks other than mallards (mean 20 days), and other waterbirds (mean 22 days) occurring later.

Waterbird use of Emiquon Preserve during spring was similar across foraging guilds (F = 0.82, P = 0.505) and years (F = 1.06, P = 0.510). Total waterbird use of Emiquon Preserve during spring migration averaged 2,495,579 ± 320,044 UDs (mean 1,445 ± 177 UD/ha; Table 1). Spring dabbling duck use-days averaged 631,130 ± 73,675 UDs with a density of 371 ± 50 UDs/ha (Fig. 2). Dabbling ducks other than mallards averaged 478,718 UDs (mean 285 ± 45 UDs/ha). Spring diving duck densities averaged 594,673 UDs (mean 364 ± 79 UDs/ha) and other waterbirds averaged 631,546 UDs (mean 362 ± 87 UDs/ha). Total use-days were less during spring than autumn migrations for most species and foraging guilds; however, lesser scaup and lesser snow geese increased use-days during spring by 16.5 and 129.5 times, respectively (Table 1).

Waterfowl behavior

During springs 2008–2013, we documented >24,000 behavior observations of dabbling and diving ducks at Emiquon Preserve. Foraging was the most common activity (mean 46%) and it varied by cover type (F = 17.27, P < 0.001), feeding guild (F = 271.59, P < 0.001), and sex (F = 4.17, P = 0.041; Table 4). Dabbling ducks (mean 62%) spent more of their time feeding than diving ducks (mean 27%). Across feeding guilds, males foraged less (mean 44%) than females (mean 48%), although the effect size was small. Across sex and feeding guilds, ducks foraged most in non-persistent emergent vegetation (mean 51%), aquatic bed (mean 51%), and hemi-marsh (mean 49%), followed by open water (mean 20%). Resting (mean 28%), locomotion (mean 18%) and other activities (mean 8%) accounted for the remainder of time spent engaged in behaviors (Table 4).
Table 4

Percent of individuals engaged in behaviors by foraging guild and habitat type at the Emiquon Preserve in Fulton County, Illinois, USA during spring migration 2008–2013

 

Behavior

Foraging

Resting

Locomotion

Other

Dabbling ducks

61.8

13.2

17.2

7.8

 Aquatic bed

69.2

6.3

18.0

6.4

 Hemi-marsh

61.4

17.8

8.5

12.3

 Non-persistent emergent

52.8

21.1

17.1

9.0

 Open water

Diving ducks

27.1

46.0

17.9

9.0

 Aquatic bed

29.0

41.7

20.6

8.6

 Hemi-marsh

40.9

29.0

19.8

10.2

 Non-persistent emergent

28.8

54.7

11.5

5.0

 Open water

19.5

57.5

13.3

9.7

All ducks

46.3

27.9

17.5

8.3

Brood observations

In 2008, the number of broods detected was similar across species and in total (P > 0.314) between flush counts (mean 31 ± 7) and passive point counts (mean 29 ± 2), except for American coot, where detections during flush counts were 7.5 ± 0.5 broods/survey greater than during passive surveys. Thus, we used passive brood surveys as an index of brood abundance for the entirety of the study period as American coots comprised only a small proportion of broods surveyed during the study.

We recorded 683 observations of waterbird broods in spring and summer at Emiquon Preserve during 2008–2013. Waterbirds began breeding at Emiquon Preserve during the first spring following flooding and densities of broods remained constant during our study period (mean 13.9 ± 1.5, range 0.0–29.5 broods/km2/survey; F = 1.44, P = 0.296). Wood ducks (mean 9.1 broods/km2/survey), Canada geese (mean 1.3 broods/km2/survey; Branta canadensis), mallard (mean 1.3 broods/km2/survey), and American coot (mean 1.1 broods/km2/survey) were the most abundant species observed with broods and densities of all species remained constant during our study period (P > 0.063). Although less common, sightings of blue-winged teal, hooded merganser (Lophodytes cucullatus), pied-billed grebe (Podilymbus podiceps), ruddy duck, Illinois-threatened black-crowned night heron (Nycticorax nycticorax), and the Illinois-endangered common gallinule (Gallinula galeata) broods were also recorded during our surveys. The mean age of broods increased by month during the spring–summer observation periods across the six years of study (F = 61.13, P = 0.016), indicating that some broods were surviving to fledge. However, average number of young per brood declined between June (mean 7.2 young/brood) and August (mean 3.5 young/brood) each year (F = 275.2, P = 0.034). The average number of young per brood did not vary across years (F = 1.52, P = 0.284).

Discussion

Waterbird response

Waterbird use in autumn 2007 was exceptional and unexpected given the relatively small wetland size and lack of extensive aquatic vegetation at the beginning of restoration. During this initial autumn, shallow water inundated the center of the historical Thompson Lake basin and flooded non-persistent vegetation (e.g., old-field vegetation) and crop stubble, where seeds and tubers were likely plentiful (Hine et al., 2016). By late 2009, the historical basins of Flag and Thompson Lakes had completely filled and water began spanning out across the former floodplain into upland vegetation and deeply inundating the shallow-marsh habitats created in 2007 (Hine et al., 2016). The obligate aquatic bed plant community that developed quickly at Emiquon Preserve (described in Hine et al., 2016) provided habitat to greater densities of many waterbird species than any other lake or wetland in the region, despite mostly passive restoration practices without supplemental planting of native aquatic species. Given the densities of waterbirds using Emiquon Preserve during the initial years of restoration and the capacity for various processes to move seeds from one wetland to another, the re-introduction of plant species through natural means seems likely (Reynolds et al., 2015).

We did not track waterbird use at a true control site (e.g., an actively farmed drainage and levee district), as aerial survey data indicate limited use of agricultural fields by most waterbirds in our region during early and mid-autumn (INHS, unpublished data). Instead, we used a more informative comparison of waterbird densities from other regional lakes and wetlands across a continuum of river connectivity and cover types (Sparks, 1995; Stafford et al., 2007). Most of these reference sites had full or partial hydrological connections to the Illinois River which allowed flooding during the spring and summer and prevented growth or maturation of emergent and submersed aquatic vegetation. Moreover, input of flood waters introduces common carp (Cyprinus carpio) and other fish, which can increase turbidity, consume vegetation, and decrease wetland foraging habitat quality for waterbirds (Ivey et al., 1998; Bowyer et al., 2005; Bajer et al., 2009). Surveyed lakes and wetlands ranged from an open connection (e.g., Peoria Lake) to predominantly hydrologic isolation (e.g., Emiquon Preserve) which resulted in different vegetation communities, depending on timing and frequency of flooding by the Illinois River (Bellrose et al., 1983; Sparks, 1995). Despite only 4 survey sites in the IRV typically containing emergent marsh and aquatic bed vegetation during autumn, waterbird use-day densities were 40% greater in those sites than in sites dominated by moist-soil and 250% greater than sites dominated by open water without vegetation. Notably, two of the other three sites in addition to Emiquon Preserve containing abundant aquatic vegetation were restored floodplains with original drainage and levee district levees intact, and all three sites were hydrologically isolated from the Illinois River and not subjected to flooding during the growing-season.

We made direct comparisons with the 931-ha south pool of the adjacent CNWR, which has a partial river connection (i.e., floodwater only enters wetlands during moderate and major flood events) and is typically managed intensively for moist-soil vegetation and flooded shallowly in autumn to provide high-quality habitat for migrating dabbling ducks (Bowyer et al., 2005). Hagy et al. (2012) documented diverse vegetation communities dominated by seed- and tuber-producing moist-soil vegetation on CNWR during 2012 which produced more than 7 million duck energy days. Although habitat conditions on CNWR were rated excellent during autumn 2012 and CNWR is considered as a model foraging habitat in the region, use-day densities of all waterbirds were still 11% greater on Emiquon Preserve (INHS, unpublished data). Although some available data indicate that energetic carrying capacity for waterbirds may be greater on a per unit basis in moist-soil wetlands than emergent marsh habitats (Soulliere et al., 2007, p. 34; Straub, 2008), we note greater use of Emiquon Preserve than CNWR by several species and guilds regardless of conditions on CNWR. Moreover, CNWR provided extensive spatial sanctuary (i.e., almost no human disturbance) and Emiquon Preserve provided only temporal sanctuary for waterbirds (i.e., periodic human disturbance from boating and hunting; Hagy et al., 2016) which further illustrates strong selection tendencies of several species and guilds for emergent marsh vegetation regardless of disturbances.

Wetland management activities on CNWR resulted in some years with high-quality foraging habitat and some years with limited drawdowns and low-quality forage that represented a range of foraging habitat quality for evaluation of different waterbird species as indicators. Regardless of foraging habitat quality on CNWR, diving ducks, total waterbirds, other waterbirds, American coots, canvasback, American wigeon, northern shoveler, and blue-winged teal use of Emiquon Preserve exceeded that of CNWR during our study and would be useful indicators of emergent marsh vegetation communities at Emiquon. However, mallard, northern pintail, American green-winged teal, dabbling ducks, and other taxa showed increased use of CNWR when moist-soil vegetation was abundant and foraging habitat quality was high in 2012 and 2013, and these taxa may not be the best indicators of emergent marsh and aquatic bed vegetation communities. Thus, generalist species, such as mallards, may be better indicators of relative habitat quality, including multiple factors such as sanctuary conditions and forage quality. Conversely, taxa with more specialized diet and habitat requirements, such as the herbivorous canvasback and American coot, may be useful to indicate aquatic vegetation communities.

Emiquon Preserve’s extensive aquatic vegetation communities, which are uncommon in the IRV, provide submersed aquatic vegetation, floating-leaf aquatic vegetation, invertebrate communities, moist-soil plants, and abundant hemi-marsh structure which are used by a wide variety of generalist (e.g., mallard) and specialist (e.g., American coot) species. For instance, blue-winged teal (659,503 UDs) and gadwall (607,453 UDs) use in 2010 was the highest estimated from aerial inventories in the IRV (INHS, unpublished data). Furthermore, northern pintail and American green-winged teal use at Emiquon Preserve in 2011 was the highest recorded at a single location in the IRV since aerial inventories began in 1948 (INHS, unpublished data). As continental population estimates of northern pintail have been below the North American Waterfowl Management Plan goal since 1976 (5.6 million; Zimpfer et al., 2012), it is noteworthy to observe such response to wetland restoration. Mallard, the most numerous duck species in North America, comprised 35% of total waterbird use-days at the nearby south pool of CNWR during our study period. Conversely, Emiquon Preserve supported a more diverse waterfowl community as mallards comprised only 11% of total waterbird use-days during or study period. Limited response and even some avoidance tendencies of mallards and other species (e.g., Canada geese) at Emiquon Preserve were likely due to their use of moist-soil and other wetland types instead of aquatic bed vegetation communities (Stafford et al., 2007, 2010). Concurrent with our aerial surveys of waterbirds, Hine et al. (2016) observed a rapid expansion of persistent emergent vegetation [e.g., cattails (Typha spp.)] and aquatic bed [e.g., submersed (i.e., Ceratophyllum dermersum, Myriophyllum spicatum) and floating-leaf aquatic plants (i.e., Nelumbo lutea, Potamogeton nodosus)] vegetation communities which produce foods typically consumed by species such as American coots and gadwall rather than mallard. For example, American coots exhibited one of the most pronounced responses to restoration. American coot use in autumn 2009 (4,249,563 UDs) was the highest observed for any surveyed location since the inception (1948) of aerial inventories in the IRV (INHS, unpublished data). The differing responses among species and foraging guilds (e.g., dabbling ducks vs. other waterbirds) demonstrate that responses of individual species or groups and diversity may be a better indicator of habitat conditions than metrics summing use and density across these groups as the latter might mask functional differences in value across species or guilds given differing life history characteristics (e.g., herbivore vs. granivore).

Use-days during autumn greatly exceeded those during spring for most species and foraging guilds. On average, total waterbird use during autumn was approximately twice that during spring, but use-days of dabbling ducks, other dabbling ducks, and other waterbirds were more than four times greater in autumn than spring. Spring migration typically advances faster than autumn migration, as most waterbird species move northward quickly following ice-out of aquatic habitats due to life-history strategies (Havera, 1999; Baldassarre & Bolen, 2006). Notably, lesser scaup and lesser snow geese increased their use of Emiquon Preserve during spring consistently across years. Lesser scaup are a species of conservation concern and recent research has indicated declining energy reserves in females and low-quality habitat conditions during spring migration (Afton & Anderson, 2001; Anteau & Afton, 2008, 2009). Although Emiquon Preserve accounted for only 7% of the total diving duck use-days during springs 2012–2014 when INHS aerial survey data were available, redhead (Aythya americana; 19%), ruddy duck (17%), common merganser (Mergus merganser; 31%), common goldeneye (Bucephala clangula; 25%), bufflehead (40%), and hooded merganser (Lophodytes cucullatus; 84%) use-days accounted for more than any other survey location during spring in the IRV, and use by these species was greater than expected compared to wetland availability in the IRV. Emiquon Preserve seems to be an important spring stopover site for several species of diving ducks during spring, but additional research is needed to determine if birds foraging on Emiquon Preserve are acquiring food resources such that they maintain a positive energy balance (Anteau & Afton, 2011).

Although use-days and density of waterbirds during autumn and spring at Emiquon Preserve seem to be suitable indicators of restoration success and site value for waterbirds, behavioral data can reveal more information about the underlying mechanisms driving site use (Johnson, 2007). For instance, foraging rates of migratory birds may predict breeding phenology (Arzel & Elmberg, 2015), food availability (Loegering & Fraser, 1995; Johnson, 2000), and overall habitat quality (Lyons, 2005). Consistent with previous research, behavioral observations revealed dabbling ducks spent most of their time feeding during autumns 2007–2009 (Hine et al., 2013) and springs 2008–2013 (Paulus, 1988). Although we lack food habits data of waterfowl using Emiquon Preserve, aquatic plants, an abundant food source at Emiquon Preserve, are usually characterized by high water and fiber content and lower true metabolizable energy than seeds, tubers, or invertebrates (Dugger et al., 2007; Coluccy et al., 2015). Gadwall diets in Louisiana consisted almost entirely (95%) of aquatic vegetation and algae, and consequently, they spent 80% of their time during the day feeding to meet nutrient requirements (Paulus, 1984). Thus, our observations illustrate the use of Emiquon Preserve as a foraging habitat of ducks during migration periods and future management actions should balance needs of providing quality foraging habitat with other functions (e.g., providing recreation opportunities; Hagy et al., 2016).

Despite substantial changes in wetland area and habitat composition during our study period (Hine et al., 2016), average waterbird brood density and size did not change across years. Average brood size declined during each summer observation period, but this was to be expected as previous studies have documented most duckling mortality occurs during the first 2 weeks after hatch (Weik & Malecki, 1999; Gendron & Clark, 2002; Hoekman et al., 2004; Yetter et al., 2009). Age classes of broods increased throughout the spring–summer observation periods indicating that Emiquon Preserve provided brood-rearing habitat capable of sustaining young waterbirds to fledging. Anecdotally, we noted declines in waterbird brood observations in 2013, which coincided with high-water conditions following a slight reduction in wetland area in 2012 during a drought. Heavy rainfall and flood waters from the Illinois River in April 2013 at the beginning of the nesting season may have negatively impacted waterbird production through displacement, nest inundation, or rapid changes in habitat availability and distribution. If the declines in waterbird productivity observed in 2013 continue concurrent with sustained high-water levels, management may be needed to increase dense emergent and hemi-marsh vegetation communities which provide habitat to brood-rearing waterbirds.

Despite the rapid colonization of Emiquon Preserve by migrating and nesting waterbirds, we predict continued declines in waterbird indicators and productivity in the future if water levels remain stable facilitating a transition into the lake-marsh stage (van der Valk & Davis, 1978). The intermediate disturbance hypothesis predicts greatest species diversity following a disturbance, as succession moves biota back towards the climax community. Waterbird density and diversity is often greatest in wetlands without stabilized water levels where periodic drying creates dynamic vegetation communities and microhabitats (Kingsford et al., 2004). At either ends of the hydrologic spectrum (i.e., ephemeral wetland vs. permanent wetland; Cowardin et al., 1979), reduced or frequent variation reduces biotic diversity; however, the type and intensity of disturbance can be a significant factor influencing species diversity in aquatic systems (Bornette et al., 1998). Although the waterbird response to restoration activities at Emiquon Preserve to date has been exceptional, use by some guilds and species is beginning to decline concurrent with changes in vegetation communities and water stabilization as conditions depart from the “intermediate” stage of disturbance and moves towards the lake marsh climax community (Hine et al., 2016).

Indicators and acceptable ranges

Typically, acceptable ranges for indicators would come from monitoring data collected at a site prior to a management action or degradation, reference sites, historical accounts from that and other similar sites prior to anthropogenic modifications, or simply target ranges based on expert opinion. However, in wetland restorations, pre-degradation data will likely be unavailable and reference sites may also be unavailable or collection of those data unaffordable. Thus, published ranges of indicators from similar restoration projects or habitats may be the only feasible baselines available and it is important for researchers to publish results from “monitoring” studies in order to build a baseline in the literature for use in ecological report cards and evaluation approaches (Tierney et al., 2009b). Ultimately, indicators should be easily quantifiable, cost effective to evaluate, robust to short-term variability and measurement errors, and relatable to a wide variety of stakeholders, including the general public (Tierney et al., 2009a). Based on monitoring the waterbird community at Emiquon Preserve and extensively in the surrounding region post restoration, we proposed target ranges for several indicators within several key ecological attributes for possible use at Emiquon Preserve in the future and other similar wetland restorations in the region (Table 5; Parrish et al., 2003).
Table 5

Potential acceptable ranges for waterbird indicators of two key ecological attributes based on the 50th and 80th percentiles of data from Emiquon Preserve in Fulton County, Illinois, USA and other INHS survey locations within the Illinois River valley during autumn and spring 2007–2013

Key Ecological Attribute

Indicator

Example Acceptable Ranges

Good (80th percentile)

Fair (80th–50th percentile)

Poor (<50th percentile)

Waterbird habitat quality

Autumn dabbling duck use-days

Ranking 1–5 (>1,132 UD/ha)

Ranking 6–12 (289–1,131 UD/ha)

Ranking <12 (<289 UD/ha)

Autumn other dabbling duck use-days

Ranking 1–5 (>493 UD/ha)

Ranking 6–12 (88–492 UD/ha)

Ranking <12 (<88 UD/ha)

Autumn other waterbird use-days

Ranking 1–5 (>110 UD/ha)

Ranking 6–12 (37–110 UD/ha)

Ranking <12 (<37 UD/ha)

Autumn diving duck use-days

Ranking 1–5 (>47 UD/ha)

Ranking 6–12 (8–47 UD/ha)

Ranking <12 (<8 UD/ha)

Autumn gadwall use-days

Ranking 1–5 (>104 UD/ha)

Ranking 6–12 (18–104 UD/ha)

Ranking <12 (<18 UD/ha)

Autumn American coot use-days

Ranking 1–5 (>88 UD/ha)

Ranking 6–12 (12–88 UD/ha)

Ranking <12 (<12 UD/ha)

Spring diving duck use-days

Ranking 1–10 (>137 UD/ha)

Ranking 11–30 (46–137 UD/ha)

Ranking <30 (<46 UD/ha)

Spring dabbling duck use-days

>486 UD/ha

486–376 UD/ha

<376 UD/ha

Spring other waterbird use-days

>469 UD/ha

469–346 UD/ha

<346 UD/ha

We suggest potential indicators because of their relative ease to measure, rapid response to changing conditions, relatability to broad audiences, availability of historical and comparative data, relevance to stakeholders, and consistency over our 7 years of monitoring at Emiquon Preserve during restoration (Lake, 2001). Because data from the period prior to the draining of Thompson and Flag Lakes, which comprised most of the aquatic habitat at Emiquon Preserve, were unavailable and response to restoration activities by waterbirds exceeded densities in most other aquatic habitat in the region including similar restorations (Bajer et al., 2009; Tables 2, 3), we suggest acceptable ranges based on the first 7 years of monitoring post restoration relative to other wetlands and lakes in the region. While our acceptable ranges are based on observational data, INHS-surveyed lakes and wetlands cover a wide range of cover types and locations that were representative of available habitats in the region.

We suggest several guilds (e.g., dabbling duck, diving duck) and species (e.g., gadwall, American coot) that may be reliable indicators of emergent and aquatic bed vegetation communities and high-quality foraging habitats. As annual variation in count frequency, migration timing, and other factors may affect waterbird counts in unpredictable ways, we suggest using use-days as a metric robust to variation. We also suggest that use-day estimates in spring and/or autumn may be more appropriate depending on the conservation targets and objectives. We found that rankings relative to reference sites were a metric even more robust to temporal variation than raw use-days (Johnson, 1980), but we acknowledge that most restoration projects may not have data available from alternative habitats or reference sites within their region and collecting those data may be cost prohibitive. Thus, we presented rankings and use-day ranges which may be helpful depending on the data available to restoration practitioners.

Conclusions

Several reviews have emphasized difficulties in using bird populations as indicators of biologic systems (Morrison, 1986; Temple & Wiens, 1989; Niemi et al., 1997), but we found that bird response to floodplain wetland restoration was exceptional and species with diets typically dominated by aquatic and other natural vegetation selected Emiquon Preserve over other available habitats in the IRV. Waterbirds are excellent indicators of ecosystem changes when long-term and large-scale data sets are available which help to overcome pitfalls associated with monitoring short-term changes in populations at particular sites without a baseline (Stolen et al., 2005; Bajer et al., 2009). We advocate ranked waterbird site use over time compared to other nearby and available habitats as an indicator of wetland quality and restoration success. Use of ranked data alleviate several biases including subjective quantification of availability and allow hierarchical interpretation of site use or habitat selection models (Johnson, 1980).

We cannot overemphasize the regional importance of Emiquon Preserve to migratory waterbirds, especially when use by some species or guilds were greater in most years than any other aerially surveyed location in the IRV. However, declining trends observed in waterbird abundance and diversity since 2009 raises significant concern regarding habitat conditions at Emiquon Preserve. Eventually, stable and high-water levels may result in declining productivity and foraging habitat quality for wildlife and management actions (e.g., drawdowns) may be necessary to reset the wetland ecosystem. As such, management can be controversial when diverse stakeholder groups are involved (Sparks et al., 2016), and we recommend engaging the regional conservation community and general public with a user-friendly evaluation process (e.g., ecosystem integrity “report card”; Harwell et al., 1999) to gain broad-based support for management activities when metrics fall outside of acceptable ranges (Tierney et al., 2009a).

Notes

Acknowledgments

We thank D. Blodgett, J. Walk, J. Beverlin, M. Lemke, T. Hobson, and others at the Nature Conservancy for supporting this project and allowing access to Emiquon Preserve; M. Cruce of Cruce Aviation for providing flight services; and several anonymous reviewers and guest editors M. Lemke, M. Lemke, and J. Walk for their helpful suggestions which have improved this manuscript. We thank The Nature Conservancy; Illinois Department of Natural Resources through Federal Aid in Wildlife Restoration; and the Illinois Natural History Survey of the Prairie Research Institute at the University of Illinois at Urbana-Champaign for providing funding and in-kind support.

References

  1. Afton, A. D. & M. G. Anderson, 2001. Declining scaup populations: a retrospective analysis of long-term population and harvest survey data. Journal of Wildlife Management 65: 781–796.CrossRefGoogle Scholar
  2. Altmann, J., 1974. Observational study of behavior: sampling methods. Behavior 49: 227–267.CrossRefGoogle Scholar
  3. Anteau, M. J. & A. D. Afton, 2008. Diet shifts of lesser scaup are consistent with the spring condition hypothesis. Canadian Journal of Zoology 84: 779–786.CrossRefGoogle Scholar
  4. Anteau, M. J. & A. D. Afton, 2009. Lipid reserves of lesser scaup (Aythya affinis) migrating across a large landscape are consistent with the spring condition hypothesis. Auk 126: 873–883.CrossRefGoogle Scholar
  5. Anteau, M. J. & A. D. Afton, 2011. Lipid catabolism of invertebrate predator indicates widespread wetland ecosystem degradation. PLoS One 6: e16029. doi: 10.1371/journal.pone.0016029.CrossRefPubMedPubMedCentralGoogle Scholar
  6. Arzel, C. & J. Elmberg, 2015. Time use and foraging behavior in pre-breeding dabbling ducks Anas spp. in sub-arctic Norway. Journal of Ornithology 156: 499–513.CrossRefGoogle Scholar
  7. Austin, J. E., T. K. Buhl, G. R. Guntenspergen, W. Norling & H. T. Sklebar, 2001. Duck populations as indicators of landscape condition in the Prairie Pothole Region. Environmental Management and Assessment 69: 29–47.CrossRefGoogle Scholar
  8. Bajer, P. G., G. Sullivan & P. W. Sorensen, 2009. Effects of a rapidly increasing population of common carp on vegetative cover and waterfowl in a recently restored Midwestern shallow lake. Hydrobiologia 632: 235–245.CrossRefGoogle Scholar
  9. Baldassarre, G. A., S. L. Paulus, A. Tamisier & R. D. Titman, 1988. Workshop summary: techniques for timing activity of wintering waterfowl. In Weller, M. W. (ed.), Waterfowl in Winter. University Minnesota Press, Minneapolis: 181–188.Google Scholar
  10. Baldassarre, G. A. & E. G. Bolen, 2006. Waterfowl ecology and Management, 2nd ed. Krieger Publishing Company, Malabar, FL.Google Scholar
  11. Beatty, W. S., E. B. Webb, D. C. Kesler, A. H. Raedeke, L. W. Naylor & D. D. Humburg, 2014. Landscape effects on mallard habitat selection at multiple spatial scales during the non-breeding period. Landscape Ecology 29: 989–1000.CrossRefGoogle Scholar
  12. Bellrose, F. C., S. P. Havera, F. L. Paveglio, Jr. & D. W. Steffeck, 1983. The fate of lakes in the Illinois River Valley. Illinois Natural History Survey Biological Notes 119.Google Scholar
  13. Blumstein, D. T., E. Fernandez-Juricic, P. A. Zollner & S. C. Garity, 2005. Inter-specific variation in avian responses to human disturbances. Journal of Applied Ecology 42: 943–953.CrossRefGoogle Scholar
  14. Bornette, G., C. Amoros & N. Lamouroux, 1998. Aquatic plant diversity in riverine wetlands: the role of connectivity. Freshwater Biology 39: 267–283.CrossRefGoogle Scholar
  15. Bowyer, M. W., J. D. Stafford, A. P. Yetter, C. S. Hine, M. M. Horath & S. P. Havera, 2005. Moist-soil plant seed production for waterfowl at Chautauqua National Wildlife Refuge, Illinois. American Midland Naturalist 154: 331–341.CrossRefGoogle Scholar
  16. Callicutt, J. T., H. M. Hagy & M. L. Schummer, 2011. The food paradigm: a review of autumn–winter food use by North American dabbling ducks (1900–2009). Journal of Fish and Wildlife Management 2(1): 29–40. doi: 10.3996/102010-JFWM-038.CrossRefGoogle Scholar
  17. Coluccy, J. M., M. V. Castelli, P. M. Castelli, J. W. Simpson, S. R. McWilliams & L. Armstrong, 2015. True metabolizable energy of American black duck foods. Journal of Wildlife Management 79: 344–348.CrossRefGoogle Scholar
  18. Cowardin, L. M., V. Carter, F. C. Golet & E. T. LaRoe, 1979. Classification of wetlands and deepwater habitats of the United States. U.S. Fish Wildlife Service. FWS/OBS 79/31. U.S. Government Printing Office, Washington, D.C.Google Scholar
  19. Dahl, T. E., 1990. Wetlands losses in the United States 1780s to 1980s. U.S. Department of the Interior, Fish and Wildlife Service, Washington, D.C.Google Scholar
  20. Dugger, B. D., M. L. Moore, R. S. Finger & M. J. Petrie, 2007. True metabolizable energy for seeds of common moist-soil plant species. Journal of Wildlife Management 71: 1964–1967.CrossRefGoogle Scholar
  21. Figarski, T. & L. Kajtoch, 2015. Alterations of riverine ecosystems adversely affect bird assemblages. Hydrobiologia 744: 287–296.CrossRefGoogle Scholar
  22. Gawlik, D. E., 2006. The role of wildlife science in wetland ecosystem restoration: lessons from the Everglades. Ecological Engineering 26: 70–83.CrossRefGoogle Scholar
  23. Gendron, M. & R. G. Clark, 2002. Survival of mallard and gadwall ducklings in south-central Saskatchewan. Journal of Wildlife Management 66: 170–180.CrossRefGoogle Scholar
  24. Gollop, J. B. & W. H. Marshall, 1954. A guide for aging duck broods in the field. Mississippi Flyway Council Technical Section Report.Google Scholar
  25. Hagy, H. M., A. P. Yetter, K. W. Stodola, M. M. Horath, C. S. Hine, M. P. Ward, T. J. Benson, R. V. Smith & J. D. Stafford, 2014. Stopover duration of mallards during autumn in the Illinois River valley. Journal of Wildlife Management 78: 747–752.CrossRefGoogle Scholar
  26. Hagy, H. M., M. M. Horath, A. P. Yetter, C. S. Hine & R. V. Smith, 2016. Evaluating tradeoffs between sanctuary for migrating waterbirds and recreational opportunities in a restored wetland complex. Hydrobiologia. doi: 10.1007/s10750-016-2711-0.Google Scholar
  27. Hagy, H. M., R. V. Smith, A. P. Yetter, C. S. Hine & M. M. Horath, 2012. Vegetation assessment of the south pool of Chautauqua National Wildlife Refuge. Illinois Natural History Survey Technical Report 2012(34)Google Scholar
  28. Harwell, M. A., V. Myers, T. Young, A. Bartuska, N. Gassman, J. H. Gentile, C. C. Harwell, S. Appelbaum, J. Barko, B. Causey, C. Johnson, A. McLean, R. Smola, P. Templet & S. Tosini, 1999. A framework for an ecosystem integrity report card. Bioscience 49: 543–556.CrossRefGoogle Scholar
  29. Havera, S. P., 1999. Waterfowl of Illinois: Status and Management. Illinois Natural History Survey Special Publication, Champaign: 21.Google Scholar
  30. Havera, S.P., K.E. Roat & L.L. Anderson. 2003. The Thompson Lake/Emiquon Story. Illinois Natural History Survey Special Publication 25. Champaign, IL. 40 p.Google Scholar
  31. Hine, C. S., H. M. Hagy, A. P. Yetter, M. M. Horath, R. V. Smith & J. D. Stafford, 2013. Waterbird and wetland monitoring at the Emiquon Preserve: final report 2007–2013. Illinois Natural History Survey Technical Report 2013 (20)Google Scholar
  32. Hine, C. S., H. M. Hagy, A. P. Yetter, M. M. Horath, R. V. Smith & J. D. Stafford, 2016. Response of aquatic vegetation and other wetland cover types to floodplain restoration at Emiquon Preserve. Hydrobiologia. doi: 10.1007/s10750-016-2893-5.Google Scholar
  33. Hoekman, S. T., T. S. Gabor, R. Maher, H. R. Murkin & L. M. Armstrong, 2004. Factors affecting survival of mallard ducklings in southern Ontario. Condor 106: 485–495.CrossRefGoogle Scholar
  34. Hohman, W. L., 1984. Diurnal time-activity budgets for ring-necked ducks wintering in central Florida. Proceedings of the Annual Conference of the Southeastern Association of Fish Game Commission 38: 158–164.Google Scholar
  35. Horath, M. M. & S. P. Havera, 2002. Illinois waterfowl surveys and investigations. Annual Federal Aid Performance Report W–43–R–49, Amendments 2 & 3. Illinois Natural History Survey, Havana, IL.Google Scholar
  36. Ivey, G. L., J. E. Cornely & B. D. Ehlers, 1998. Carp impacts on waterfowl at Malheur National Wildlife Refuge. Transactions of the 63rd North American Wildlife and Natural Resources Conference. Wildlife Management Institute, Washington, D.C.Google Scholar
  37. Jackson, C. R. & C. M. Pringle, 2010. Ecological benefits of reduced hydrologic connectivity in intensively developed landscapes. BioScience 60: 37–46.CrossRefGoogle Scholar
  38. Johnson, D. H., 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61: 65–71.CrossRefGoogle Scholar
  39. Johnson, M. D., 2000. Evaluation of an arthropod sampling technique for measuring food availability for forest insectivorous birds. Journal of Field Ornithology 71: 88–109.CrossRefGoogle Scholar
  40. Johnson, M. D., 2007. Measuring habitat quality: a review. Condor 109: 489–504.CrossRefGoogle Scholar
  41. Kajtoch, L., M. Zmihorski & A. Piestrzynska-Kajtoch, 2014. The goosander as potential indicator of naturalness and biodiversity in submontane river valleys of northern Carpathians. Ecological Indicators 45: 83–92.CrossRefGoogle Scholar
  42. Kaminski, R. M. & H. H. Prince, 1981. Dabbling duck activity and foraging responses to aquatic invertebrates. Auk 98: 115–126.Google Scholar
  43. King, S. L., D. J. Twedt & R. R. Wilson, 2006. The role of Wetland Reserve Program conservation efforts in the Mississippi River Alluvial Valley. Wildlife Society Bulletin 34: 914–920.CrossRefGoogle Scholar
  44. Kingsford, R. T., K. M. Jenkins & J. L. Porter, 2004. Imposed hydrological stability on lakes in arid Australia and effects on waterbirds. Ecology 85: 2478–2492.CrossRefGoogle Scholar
  45. Lake, P. S., 2001. On the maturing of restoration: linking ecological research and restoration. Ecological Management and Restoration 2: 110–115.CrossRefGoogle Scholar
  46. Linz, G. M., D. C. Blixt, D. L. Bergman & W. J. Bleier, 1996. Response of ducks to glyphosate-induced habitat alterations in wetlands. Wetlands 16: 38–44.CrossRefGoogle Scholar
  47. Littell, R. C., G. A. Milliken, W. W. Stroup, R. D. Wolfinger & O. Schabenberger, 2006. SAS for Mixed Models, 2nd ed. SAS Institute, Inc., Cary, North Carolina, USA.Google Scholar
  48. Loegering, J. P. & J. D. Fraser, 1995. Piping plover survival in different brood-rearing habitats. Journal of Wildlife Management 59: 646–655.CrossRefGoogle Scholar
  49. Lyons, J. E., 2005. Habitat specific foraging of Prothonotary Warblers: deducing habitat quality. Condor 107: 41–49.CrossRefGoogle Scholar
  50. Mills, H. B., W. C. Starrett & F. C. Bellrose, 1966. Man’s effect on the fish and wildlife of the Illinois River. Illinois Natural History Survey Biological Notes 57.Google Scholar
  51. Moore, M., S. P. Romano & T. Cook, 2010. Synthesis of Upper Mississippi River system submersed and emergent aquatic vegetation: past, present, and future. Hydrobiologia 640: 103–114.CrossRefGoogle Scholar
  52. Morrison, M. I., 1986. Bird populations as indicators of environmental change. In Johnston, R. F. (ed.), Current Ornithology. Plenum, New York: 429–451.CrossRefGoogle Scholar
  53. North American Waterfowl Management Plan (NAWMP), 2012. North American waterfowl management plan: people conserving waterfowl and wetlands [available on internet at http://www.dgif.virginia.gov/wildlife/waterfowl/north-american-waterfowl-management-plan-2012.pdf]
  54. Niemi, G. J., J. M. Hanowski, A. R. Lima, T. Nicholls & N. Wiland, 1997. A critical analysis on the use of indicator species in management. Journal of Wildlife Management 61: 1240–1252.CrossRefGoogle Scholar
  55. O’Neal, B. J., J. D. Stafford & R. P. Larkin, 2012. Stopover duration of fall-migrating dabbling ducks. Journal of Wildlife Management 76: 285–293.CrossRefGoogle Scholar
  56. Pagano, A. M. & T. W. Arnold, 2009. Estimating detection probabilities of waterfowl broods from ground-based surveys. Journal of Wildlife Management 73: 686–694.CrossRefGoogle Scholar
  57. Paracuellos, M., 2006. How can habitat selection affect the use of a wetland complex by waterbirds? Biodiversity and Conservation 15: 4569–4582.CrossRefGoogle Scholar
  58. Parrish, J. D., D. P. Braun & R. S. Unnasch, 2003. Are we conserving what we say we are? Measuring ecological integrity within protected areas. BioScience 53: 851–860.CrossRefGoogle Scholar
  59. Paulus, S. L., 1984. Activity budgets of nonbreeding gadwalls in Louisiana. Journal of Wildlife Management 48: 371–380.CrossRefGoogle Scholar
  60. Paulus, S. L., 1988. Time-activity budgets of nonbreeding Anatidae: a review. In Weller, M. W. (ed.), Waterfowl in Winter. University Minnesota Press, Minneapolis, MN: 135–149.Google Scholar
  61. Reynolds, C., N. A. Miranda & G. S. Cumming, 2015. The role of waterbirds in the dispersal of aquatic alien and invasive species. Diversity and Distributions 21: 744–754.CrossRefGoogle Scholar
  62. Ringelman, J. K. & L. D. Flake, 1980. Diurnal visibility and activity of blue-winged teal and mallard broods. Journal of Wildlife Management 44: 822–829.CrossRefGoogle Scholar
  63. Rumble, M. A. & L. D. Flake, 1982. A comparison of two waterfowl brood survey techniques. Journal of Wildlife Management 46: 1048–1053.CrossRefGoogle Scholar
  64. Smith, R. V., J. D. Stafford, A. P. Yetter, M. M. Horath, C. S. Hine & J. P. Hoover, 2012. Foraging ecology of fall-migrating shorebirds in the Illinois River valley. PLoS One 7(9): e45121. doi: 10.1371/journal.pone.0045121.CrossRefPubMedPubMedCentralGoogle Scholar
  65. Soulliere, G. J., B. A. Potter, J. M. Coluccy, R. C. Gatti, P. W. Brown, C. L. Roy Nielsen, M. W. Eichholz & D. R. Luukkonen, 2007. Upper Mississippi River and Great Lakes Region Joint Venture Waterfowl Habitat Conservation Strategy. U.S. Fish and Wildlife Service, Fort Snelling, Minnesota.Google Scholar
  66. Sparks, R. E., 1984. The role of contaminants in the decline of the Illinois River: implications for the Upper Mississippi. In Wiener, J. G., R. V. Anderson & D. R. McConville (eds), Contaminants in the Upper Mississippi River. Butterworth Publishers, Stoneham, MA: 25–66.Google Scholar
  67. Sparks, R. C., 1995. Need for ecosystem management of large rivers and their floodplains. Bioscience 45: 168–182.CrossRefGoogle Scholar
  68. Sparks, R. E., K. D. Blodgett, A. F. Casper, H. M. Hagy, M. L. Lemke, A. F. M. Velho & L. C. Rodrigues, 2016. Why experiment with success? Opportunities and risks in applying assessment and adaptive management to the Emiquon floodplain restoration project. Hydrobiologia. doi: 10.1007/s10750-016-2785-8.
  69. Stafford, J. D., M. M. Horath, A. P. Yetter, C. S. Hine & S. P. Havera, 2007. Wetland use by mallards during spring and fall in the Illinois and Central Mississippi River Valleys. Waterbirds 30: 394–402.CrossRefGoogle Scholar
  70. Stafford, J. D., M. M. Horath, A. P. Yetter, R. V. Smith & C. S. Hine, 2010. Historical and contemporary characteristics and waterfowl use of Illinois River valley wetlands. Wetlands 30: 565–576.CrossRefGoogle Scholar
  71. Stolen, E. D., D. R. Breininger & P. C. Frederick, 2005. Using waterbirds as indicators in Estuarine systems: successes and perils. In. Bortone, S. A. (ed.), Estuarine Indicators. CRC Press, Boca Raton, FL: 409–422Google Scholar
  72. Straub, J. N., 2008. Energetic carrying capacity of habitats used by spring-migrating waterfowl in the Upper Mississippi River and Great Lakes region. M.S. Thesis, The Ohio State University, Columbus, OH.Google Scholar
  73. Suding, K. N., K. L. Gross & G. R. Houseman, 2004. Alternative states and positive feedbacks in restoration ecology. Trends in Ecology and Evolution 19: 46–53.CrossRefPubMedGoogle Scholar
  74. Temple, S. A. & J. A. Wiens, 1989. Bird populations and environmental changes: can birds be bio-indicators? American Birds 43: 260–270.Google Scholar
  75. The Nature Conservancy (TNC). 2006. Key attributes and indicators for Illinois River conservation targets at The Nature Conservancy’s Emiquon Preserve. The Nature Conservancy, Peoria, IL. USA. [available on internet at www.uis.edu/emiquon/wp-content/uploads/sites/78/2013/04/KEA_Report_and_Analysis.pdf]
  76. Tierney, G. L., D. Faber-Langendoen, B. R. Mitchell, W. G. Shriver & J. P. Gibbs, 2009a. Monitoring and evaluating the ecological integrity of forest ecosystems. Frontiers in Ecology and the Environment 7: 308–316.CrossRefGoogle Scholar
  77. Tierney, G. L., B. Mitchell, K. Miller, J. Comiskey, A. Kozlowski & D. Faber-Langendoen, 2009b. Long-term forest monitoring protocol: northeast temperate network. Natural Resource Report NPS/NETN/NRR–2009/117. National Park Service, Fort Collins, CO.Google Scholar
  78. Tockner, K. & J. A. Stanford, 2002. Riverine flood plains: present state and future trends. Environmental Conservation 29: 308–330.CrossRefGoogle Scholar
  79. U.S. Fish & Wildlife Service (USFWS), 2014. Waterfowl population status, 2014. U.S. Department of the Interior, Washington, D.C., USA [available on internet at http://flyways.us/sites/default/files/uploads/statusreport2014_final_7-24-14.pdf]
  80. van der Valk, A. G. & C. B. Davis, 1978. The role of seed banks in the vegetation dynamics of prairie glacial marshes. Ecology 59: 322–335.CrossRefGoogle Scholar
  81. Van Middlesworth, T. D., N. N., McClelland, G. G. Sass, A. F. Casper, T. W. Spier & M. J. Lemke, 2016. Fish community succession and biomanipulation to control two common aquatic ecosystem stressors during a large-scale floodplain lake restoration. Hydrobiologia. doi: 10.1007/s10750-016-2696-8.
  82. Webb, E. B., L. M. Smith, M. P. Vrtiska & T. G. Lagrange, 2010. Effects of local and landscape variables on wetland bird habitat use during migration through the Rainwater Basin. Journal of Wildlife Management 74: 109–119.CrossRefGoogle Scholar
  83. Weik, A. P. & R. A. Malecki, 1999. Survival of mallard ducklings, broods, and hens in western New York. Northeast Wildlife 54: 1–11.Google Scholar
  84. Weller, M. W. & L. H. Fredrickson, 1974. Avian ecology of a managed glacial marsh. Living Bird 12: 269–291.Google Scholar
  85. Yetter, A. P., J. D. Stafford, C. S. Hine, M. W. Bowyer, S. P. Havera & M. M. Horath, 2009. Nesting biology of mallards in west-central Illinois. Illinois Natural History Survey Bulletin 39: 1–38.Google Scholar
  86. Zar, J. H., 2009. Biostatistical Analysis, 5th ed. Prentice Hall, Upper Saddle River, NJ.Google Scholar
  87. Zimpfer, N. L., W. E. Rhodes, E. D. Silverman, G. S. Zimmerman & K. D. Richkus, 2012. Trends in duck breeding populations, 1955–2012. U.S. Fish and Wildlife Service Administrative Report.[available on internet at http://www.fws.gov/migratorybirds/Newreportspublications/PopulationStatus/Trends/2012_Trend_Report_final.pdf].

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Heath M. Hagy
    • 1
  • Christopher S. Hine
    • 1
  • Michelle M. Horath
    • 1
  • Aaron P. Yetter
    • 1
  • Randolph V. Smith
    • 1
    • 2
  • Joshua D. Stafford
    • 1
    • 3
  1. 1.Illinois Natural History Survey, Forbes Biological Station – Bellrose Waterfowl Research CenterUniversity of Illinois at Urbana-ChampaignHavanaUSA
  2. 2.Illinois Department of Natural ResourcesSpringfieldUSA
  3. 3.U.S. Geological Survey, South Dakota Cooperative Fish and Wildlife Research Unit, Department of Natural Resource ManagementSouth Dakota State UniversityBrookingsUSA

Personalised recommendations