Microhabitat associations of wintering birds in a Southeastern bottomland forest within the Eastern Gulf Coastal Plain of Florida

Tittensor, D. P., A. R. Baco, P. E. Brewin, M. R. Clark, M. Consalvey, S. J. Hall, A. A.
Rowden, T. Schlacher, K. I. Stocks, A. D. Rogers. 2009. Predicting global habitat
suitability for stony corals on seamounts. Journal of Biogeography 36:1111-1128.
Townsend, P. A., and S. J. Walsh. 2001. Remote sensing of forested wetlands:
application of multitemporal and multispectral satellite imagery to determine
plant community composition and structure in southeastern USA. Plant Ecology
157:129-149.
U.S. Fish & Wildlife Service. 2007. U.S. Fish & Wildlife Service, National Wetlands
Inventory. [Online.] Available at http://www.fws.gov/nwi/.
Wakeley J. S., and T. H. Roberts. 1996. Bird distributions and forest zonation in a
bottomland hardwood wetland. Wetlands 16:296-308.
Weldon, A. J., and N. M. Haddad. 2005. The effects of patch shape on indigo buntings:
Evidence for an ecological trap. Ecology 86:1422-1431.
Wear, D. N., and J. G. Greis, eds. 2002. Southern forest resource assessment. Gen.
Tech. Rep. SRS-53. Asheville, NC: U.S. Department of Agriculture, Forest
Service, Southern Research Station.
Wheelwright, N. T. 1986. The diet of American Robins: an analysis of U. S. biological
survey records. Auk 103:710-725.
Wilkins, H. D. 2001. The winter foraging ecology of Yellow-bellied Sapsuckers,
Sphyrapicus varius, in east-central Mississippi. Ph.D. dissertation, Mississippi
State University, Mississippi State.

28

Wintle, B. A., R. P. Kavanagh, M. A. McCarthy, M. A. Burgman. 2005. Estimating and
dealing with detectability in occupancy surveys for forest owls and arboreal
marsupials. Journal of Wildlife Management. 69:905-917.
Zeller, N. S., and J. A. Collazo. 1995. Abundance and distribution of overwintering
passerines in bottomland hardwood forests in North Carolina. Wilson Bulletin
107:698-708.
Yarbrough, C. G., and D. W. Johnston. 1965. Lipid deposition in wintering and
premigratory Myrtle Warblers. Wilson Bulletin 77:175-191.
Zeller, N. S., and J. A. Collazo. 1995. Abundance and distribution of overwintering
passerines in bottomland hardwood forests in North Carolina. Wilson Bulletin
107:698-708.

29

Table 1. Covariates affecting detectability and microhabitat characteristics that were
tested in an assessment of habitat selection with their designated category, name,
explanation, number of species tested, and number of times covariates were found
important.

Category Covariate name Explanation ^{Number of}
species tested
Number of
models with
QAIC = 0
Detectability Date Scaled Julian date 34 18
Time Minutes after sunrise 34 5
Temp Temperature (Celsius) 34 6
Wind Wind speed Beaufort scale 34 10
Observer Observer 34 10
Physiognomic Basal area Total basal area 22 1
Deciduous Basal area of deciduous trees 1 0
Evergreen Basal area of evergreen trees 3 0
Snags Basal area of snags 9 0
Woody debris Volume of downed woody debris 5 0
Canopy Upward densitometer readings in 4-cardinal directions 27 1
Ground Downward densitometer readings in 4-cardinal directions 22 2
Woody stems Number of woody stems < 10-cm dbh 21 4
Height Height of 3 tallest trees 6 0
Length Canopy length of 3 tallest trees 2 0
_{Leaf litter} Average depth of leaf litter measured in 4 cardinal
directions at the edge and center of the plot ^{14 2}
Mixed Index of mixed pine and non-pine forest 4 0
Number of trees Number of trees 5 0
Exposed bark calculated as the surface area of a cone from
Bark 5 0
the 3 most dominant trees
^{Number of trees > 30} Number of large trees > 30-cm dbh 1 0
cm
^{Woody debris > 30} Volume of large woody debris > 30 cm at one end 1 0
cm
Water Presence or absence of water 14 3
Floristic Oak Quercus nigra ,Q. michauxii, Q. lyrata 30 0
Holly Ilex opaca 10 2
Tupelo Mostly Nyssa aquatica, also N. sylvatica 19 6
Elm Planera aquatica, Ulmus americana 7 1
Cypress Taxodium distichum 18 0
Pine Pinus taeda, P. glabra 18 2
Cottonwood Populus 7 0
Sweetgum Liquidambar styraciflua 4 1
Maple Mostly Acer rubrum 3 1
Ash Mostly Fraxinus pennsylvanica 11 0

30

Table 1. continued.

Category Covariate name Explanation ^{Number of}
species tested
Number of
models with
QAIC = 0
Floristic Hickory Carya aquatica 20 2
Magnolia Mostly Magnolia grandiflora, also M. virginiana 1 0
Willow Mostly Salix nigra 1 0
Sycamore Platanus occidentalis 10 0
Wateroak Quercus nigra 1 0
_{Oak-gum-cypress} Quercus + Nyssa + Taxodium community where all three
are present ^{20 2}
Oak-hickory Quercus + Carya community where both are present 20 3
Mistletoe Phoradendron 2 0
Cane Cover of giant cane; Arundinaria gigantea 1 0
Palmetto Cover of saw palmetto; Serenoa repens 3 1

31

Table 2. Species grouped by covariates with their β values for each species where an
association was determined by the best models (QAIC = 0). Table 1 explains each
microhabitat characteristic group.

Category Covariate Group ^Species β (SE)
Physiognomic Woody stems ^{Winter Wren} -0.45 (0.26)
Downy Woodpecker _{-0.48 (0.20)}
Rusty Blackbird _{-1.05 (0.71)}
Hermit Thrush _{0.53 (0.53)}
Water ^{Hermit Thrush} -1.19 (0.80)
Wood Duck _{0.70 (0.41)}
Eastern Phoebe _{0.89 (0.45)}
Ground ^{American Goldfinch} -0.86 (0.61)
American Robin _{-0.93 (0.37)}
Leaf litter ^{Pine Warbler} 0.34 (0.16)
Common Grackle _{0.34 (0.21)}
Basal area ^{Blue-headed Vireo} -0.38 (0.16)
Canopy ^{Cedar Waxwing} 0.54 (0.27)
Floristic Tupelo ^{Ruby-crowned Kinglet} -0.52 (0.18)
Winter Wren _{-0.63 (0.38)}
Barred Owl _{-0.96 (0.58)}
Common Grackle _{-1.04 (0.68)}
American Goldfinch _{-1.54 (1.05)}
Black Vulture _{0.78 (0.28)}
Holly ^{Northern Cardinal} 15.58 (9.32)
Red-bellied Woodpecker _{37.76 (15.07)}
Pine ^{Yellow-rumped Warbler} -3.45 (1.88)
Golden-crowned Kinglet _{4.08 (3.22)}
Hickory ^{Wood Duck} 3.10 (1.46)
Yellow-bellied Sapsucker _{3.88 (1.23)}
Elm ^{Red-shouldered Hawk} -95.98 (85.22)
Sweetgum ^{American Goldfinch} 0.77 (0.53)
Maple ^{Hairy Woodpecker} 2.03 (0.95)
Palmetto ^{Hermit Thrush} 1.00 (0.65)
Oak-gum-cypress ^{Red-shouldered Hawk} 0.44 (0.26)
Pileated Woodpecker _{2.23 (1.28)}
Oak-hickory ^{American Crow} -0.48 (0.18)
Eastern Towhee _{-1.42 (0.75)}
Cedar Waxwing _{0.54 (0.33)}

32

Figure 1. Sites sampled in the Choctawhatchee River Basin, FL for assessing habitat
selection by wintering birds. Irregular 2-km² grids conformed to the river basin and point
transects were randomly selected from within each grid. Variations in vegetation are
more distinct in infrared photographs than in aerial photographs.

33

Figure 2. The hierarchical structure of occupancy analyses used in an assessment of
habitat selection by wintering birds in the Choctawhatchee River Basin, Florida.
Detectability covariates were tested first and incorporated within all subsequent analyses.
Physiognomic and floristic covariates were tested. Lastly, physiognomic and floristic
models were combined and compared. I only incorporated models with ∆QAIC values ≤
2.0 and models with high weights from physiognomic and floristic result into the final
analysis.

34

N^umber
iⁿ
b^est
m^odel
(^ΔQAIC
=
0⁾

25

20

15

10

5
Observed
Expected

0

Physiognomic
Microhabitat category
Floristic

Figure 3. Physiognomic and floristic characteristics were included in the best models for
a statistically similar number of bird species. There was no significant difference (Yates’
continuity corrected χ²=0.00, p=0.93, df = 1, Fig. 3) between observed and expected
frequencies of microhabitat characteristics.

35

Figure 4. Effects of physiognomic characteristics on occupancy of birds in an assessment
of habitat selection by wintering birds in the Choctawhatchee River Basin, Florida.
Modeled effects are (A) cover of palmetto on Hermit Thrushes (HETH), (B) depth of leaf
litter on Common Grackles (COGR) and Pine Warblers (PIWA), (C) the presence or
absence of water on Eastern Phoebes (EAPH), Hermit Thrushes, and Wood Ducks
(WODU), (D) canopy cover on Cedar Waxwings (CEDW), (E) total basal area on Blueheaded
Vireos (BHVI), (F) number of woody stems < 10-cm dbh on Downy
Woodpeckers (DOWO), Hermit Thrushes, Rusty Blackbirds (RUBL), and Winter Wrens
(WIWR), and (G) ground cover on American Goldfinches (AMGO) and American
Robins (AMRO). Note that palmetto is actually a floristic component, but is included
here for comparative purposes. Standard errors for each β value are included on Table 2.

36

Figure 5. Effects of floristic characteristics on occupancy in an assessment of habitat
selection by wintering birds in the Choctawhatchee River Basin, Florida. Modeled
effects are (A) tupelo on American Goldfinches (AMGO), Barred Owls (BDOW), Black
Vultures (BLVU), Common Grackles (COGR), Ruby-crowned Kinglets (RCKI) and
Winter Wrens (WIWR), (B) holly on Northern Cardinals (NOCA) and Red-bellied
Woodpeckers (RBWO), (C) pine on Golden-crowned Kinglets (GCKI) and Yellowrumped
Warblers (YRWA), (D) Sweetgum on American Goldfinches, (E) oak-gumcypress
on Pileated Woodpeckers (PIWO) and Red-shouldered Hawks (RSHA), (F)
hickory on Wood Ducks (WODU) and Yellow-bellied Sapsuckers (YBSA), (G) maple on
Hairy Woodpeckers (HAWO), (H) oak-hickory on American Crows (AMCR), Cedar
Waxwings (CEDW), and Eastern Towhees (EATO), (I) elm on Red-shouldered Hawks.
Note that although palmetto is a floristic covariate, palmetto is included with
physiognomic characteristics in Fig. 4 for comparative purposes. Standard errors for
each β value are included on Table 2.

37