Effect of Group-Selection Opening Size on Breeding Bird Habitat Use in a Bottomland Forest Page: 4 of 12
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to UNT Digital Library by the UNT Libraries Government Documents Department.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
EFFECT OF GAP SIZE ON BOTTOMLAND BIRDS
USGS BRD (U.S. Geological Survey, Biological Re-
sources Division) aluminum leg band. We also recorded
the net location of each capture within the gap grid
(i.e., gap edge, gap interior, or forest adjacent to gap).
Vegetation sampling
In July 1996, we measured the forest vegetation ad-
jacent to gaps and at control sites, using a modification
of the James and Shugart (1970) technique. We mea-
sured the vegetption at a single 0.04-ha circular plot
(hereafter, forest plot) centered on a randomly chosen
corner of each spot-map grid. Within each forest plot,
we tallied all woody stems >3 cm in diameter at breast
height (dbh) into three diameter classes (3-8 cm, 9-
23 cm, >23 cm). Basal area of hardwood and pine
stems, number of hardwood and pine stems, and num-
ber of snags (i.e., standing dead stems >3 cm dbh) also
were calculated. We used a spherical densiometer (four
samples per site) to estimate percent canopy cover and
an ocular tube (20 samples per site) to determine per-
cent ground cover (James and Shugart 1970). We de-
termined heights of the three tallest canopy trees in
each 0.04-ha forest plot using a clinometer, and their
average was used as the forest plot canopy height.
We measured understory vegetation in gaps from
mid-to-late July in 1996, 1997, and 1998, using the
technique described by Wiens (1969), as modified by
Rotenberry and Wiens (1980). We assumed that veg-
etative changes in the unharvested controls would be
minimal, so understory vegetation at control grids was
measured only once (July 1996) during the three years
of the study. We measured vegetation at nine plots
(hereafter, understory plots) in each 0.5-ha gap; seven
plots in each 0.26-ha gap; and five plots in each of the
0.13-ha gaps, 0.06-ha gaps, and control grids. In gaps,
we established understory plots 2 m from the north and
south edges, at the center, and at one-half the radius in
the east and west directions from the gap center. In
0.26-ha gaps, understory plots also were established at
one-half the radius and 450 from two randomly chosen
cardinal directions (e.g., northeast and southwest). In
0.5-ha gaps, understory plots also were created at one-
half the radius and 450 from all cardinal directions. We
established all live control grid plots within the 0.04-
ha forest plots previously described. Control understo-
ry plots were located at the center of the 0.04-ha forest
plot and 10 m from the forest plot center in each of
the four cardinal directions. At gap and control grids,
we measured the vertical distribution of understory
vegetation at sampling points 2 m east and west of each
understory plot center (e.g., 18 sampling points in 0.5-
ha gaps). We estimated vertical structure by recording
whether or not vegetation touched each 1-dm height
interval of a 2-m rod passed vertically through the veg-
etation. Plants that hit the rod were recorded as woody,
grass/sedge, forb, or slash. Where vegetation occurred
in the sampling plane above 2 in in gaps, the maximum
height of vegetation above 2 rn was recorded.From the rod data, we calculated percent horizontal
cover for each of the four vegetative types (WOOD,
FORB, GRASS, SLASH) in gap and control grids by
dividing the number of sampling points where vege-
tation intersected the rods by the total number of sam-
pling points measured. We indexed vertical structure
by calculating the mean total number of decimeters
with vegetative hits (TOTHIT) and the mean maximum
decimeter height interval contacting vegetation in gaps
(MAXHT). We calculated variation among sampling
points in vertical structure within a gap as an index of
horizontal heterogeneity. We used coefficients of var-
iation (cv) of the two vertical structure variables
(CVTOTHIT, CVMAXHT) to estimate such variation.
To estimate heterogeneity at a smaller scale (within
understory plot), we calculated a heterogeneity index
based on within-plot differences in TOTHIT and
MAXHT and averaged over an entire gap (Rotenberry
and Wiens 1980). The total hits heterogeneity index
(HIT-HI) and maximum height heterogeneity index
(MAX-HI) were defined as: HIT-HI or MAX-HI = 1
(High - Low)/I x, where High is the high value for
TOTHIT or MAX-HI within an understory plot, Low
is the low value for TOTHIT or MAX-HI within an
understory plot, and i is the mean value of TOTHIT
or MAX-HI within an understory plot. Because
MAXHT was calculated only for gaps, CVMAXHT and
MAX-HI were not computed for control grids.
Statistical analysis
We grouped bird species into three habitat-use as-
sociations (field edge, forest edge, and forest interior)
following Freemark and Collins (1992), Hamel (1992),
and Kilgo et al. (1998). Two of these references were
based on data collected from the southeastern United
States, and should more closely describe the regional
habitat associations of species detected during our
study. Further, we grouped bird species as either res-
idents or neotropical migrants (Hamel 1992). Non-
breeding migrant species (i.e., transients) were exclud-
ed from statistical analyses.
Using spot mapping data, we tested for differences
in species richness, total detections, and numbers of
detections by species, habitat association, and migra-
tory strategy among treatments (i.e., four gap sizes and
control grids) using a split-plot in time (3-yr) analysis
of covariance (ANCOVA; SAS Institute 1996). Habitat
variables measured in the forest plots were included as
covariates in the ANCOVA. We examined correlations
among forest plot habitat variables using Pearson prod-
uct-moment correlation coefficients, and the most eas-
ily measured (i.e., most likely to be measured during
timber inventories) variable of a correlated pair (P -
0.05) was retained as a covariate. Four of I 1 variables
measured in forest plots were retained for inclusion as
covariates in analyses of mapping data (Table 1). When
covariate effects were nonsignificant, they were elim-
inated. When the ANCOVA yielded a significant F sta-December 2001
1683
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Moorman, C.E. & D.C. Guynn, Jr. Effect of Group-Selection Opening Size on Breeding Bird Habitat Use in a Bottomland Forest, article, December 1, 2001; New Ellenton, South Carolina. (https://digital.library.unt.edu/ark:/67531/metadc738796/m1/4/: accessed April 25, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.