Content uploaded by George M Linz
Author content
All content in this area was uploaded by George M Linz
Content may be subject to copyright.
Wildlife Conservation Sunflower Plots and Croplands as Fall
Habitat for Migratory Birds
HEATH M. HAGY
1
Department of Biological Sciences, North Dakota State University, Stevens Hall, Fargo 58105
GEORGE M. LINZ
United States Department of Agriculture, Wildlife Services, National Wildlife Research Center, Bismarck,
North Dakota 58501
AND
WILLIAM J. BLEIER
Department of Biological Sciences, North Dakota State University, Stevens Hall, Fargo 58105
ABSTRACT.—Agricultural fields are often overlooked as post-breeding and migratory bird
habitat, even though many species use row-crop fields in the northern Great Plains. We
monitored bird use, crop and non-crop vegetation characteristics and abundance, and land
use around (#2.4 km) 35 8-ha Wildlife Conservation Sunflower Plots (WCSP) and one
commercial sunflower and non-sunflower row-crop field, paired with each WCSP, in fall 2004
and 2005. We excluded four species of blackbirds from our analysis that commonly form large
foraging flocks and ubiquitously use agricultural fields. A diverse community of migratory
birds used WCSP and commercial sunflower compared to other non-sunflower crops in
North Dakota. Both WCSP (mean 524.4 birds/ha, SE 52.7) and commercial sunflower
(mean 512.7 birds/ha, SE 51.7) harbored greater densities of birds (P ,0.01) than did
other non-sunflower row crops (mean 57.2 birds/ha, SE 51.1). Migratory birds were more
strongly associated with vegetation within fields, such as crop density (+), non-crop plant
abundance (+) and crop height (+), than surrounding land uses (0–2.4 km from WCSP). We
recommend management practices to maximize WCSP for fall bird habitat and discuss
economic considerations for WCSP implementation as a wildlife habitat / blackbird damage
management system.
INTRODUCTION
Agricultural crops provide habitat for migratory birds in the northern Great Plains (NGP)
of North America during and before migration periods. Extensive cultivation of crops has
fragmented .50%of prairies in the NGP and likely influenced bird communities in the
region (Peterjohn, 2003; Lubowski et al., 2006). Few studies have examined migratory bird
use of croplands in North America, especially during non-breeding periods, even though
croplands represent the third-largest land use (19.5%) in the United States (Lubowski et al.,
2006).
Birds migrating during fall likely select staging habitat hierarchically based on foraging
and energy demands of migration, landscape composition and weather (Johnson, 1980;
Kolasa, 1989; Kotliar and Wiens, 1990; Bergin, 1992; Moore and Simons, 1992; Moore and
Aborn, 2000). Croplands may be selected based on patch quality detectable from proximate
cues or composition of habitat complexes (Best et al., 1990, 1998; Flather and Sauer, 1996;
1
Corresponding author present address: Department of Wildlife and Fisheries, Mississippi State
University, Box 9690, Mississippi State 39762; e-mail: hhagy@cfr.msstate.edu; Telephone: (662) 325-
4739
Am. Midl. Nat. 164:119–135
119
Koford and Best, 1996; Schaaf, 2003; Galle, 2005; Pearse, 2007). Within selected landscapes,
land use diversity surrounding roost and loafing sites and vegetative characteristics within
fields likely influence field selection and subsequent use (Stone and Danner, 1980; Best et
al., 2001; Cunningham and Johnson, 2006). Some croplands within the NGP may provide
disproportionately important habitat given crop characteristics, farming practices, time of
harvest relative to other crops and associated food resources.
Birds use sunflower and other crops during migration through the NGP (Best et al., 1998;
Murphy, 2003; Hagy et al., 2007), but we know little about landscape-scale or within-field
variables influencing use (Moore et al., 1995; Peterjohn, 2003). Sunflower is used extensively
by migrating birds, especially after mid- to late-summer harvest and tilling of other row crops
reduce vegetation and vertical structure of croplands in the NGP (Schaaf, 2003; Linz et al.,
2003). In late summer and fall, an energy-rich and structurally diverse crop (Schaff, 2003)
such as sunflower may be easily recognizable as a source of cover and food for a variety of
birds (Charlet et al., 1997; Linz et al., 2004; Hagy et al., 2007).
Blackbirds [red-winged blackbirds (Agelaius phoeniceus), Brewer’s blackbird (Euphagus
cyanocephalus), common grackles (Quiscalus quiscula) and yellow-headed blackbirds
(Xanthocephalus xanthocephalus)] extensively use sunflower fields and subsequently cause
economically important damage by feeding on ripening sunflower seeds (Otis and Kilburn,
1988; Peer et al., 2003; Hagy et al., 2008). Traditional harassment and damage prevention
methods are time-intensive, expensive and negatively affect other wildlife and bird species
(Kleingartner, 2003; Linz et al., 2003). Furthermore, population declines of several
blackbird species (Brewer’s blackbird, red-winged blackbird and common grackle) since
1966 and negative public sentiment associated with lethal population control incentivize
non-lethal pest management practices (Kleingartner, 2003; Sauer et al., 2008). Blackbird
depredation of commercial sunflower can be mediated by strategically planting small
sunflower lure plots to concentrate blackbird foraging flocks away from commercial
sunflower fields (Cummings et al., 1987; Hagy et al., 2008). Additionally, lure plots may
provide stopover habitat for other migratory bird species and wildlife and reduce
disturbance of non-blackbird species using commercial sunflower fields.
Farmers contracted with the United States Department of Agricultural/Animal and Plant
Health Inspection Service’s division of Wildlife Services to plant 8 ha plots of oilseed
sunflower in 2004 (n 514) and 2005 (n 521), called Wildlife Conservation Sunflower Plots
(WCSP). WCSP were planted to lure blackbirds away from commercial sunflower fields and
provide habitat for other migratory birds and wildlife (Cummings et al., 1987). Cost-share
opportunities to plant WCSP were awarded to farmers based on historical county-wide
damage estimates, history of damage on site and presence of cattail-dominated wetlands that
blackbirds typically use as night roosts in proposed county sections. Individual farmers who
were accepted in the WCSP program determined placement within the approved county
section and planted WCSP. Hagy et al. (2008) reported guidelines for maximizing WCSP
placement to reduce blackbird damage to sunflower, but did not describe use of plots or
surrounding croplands by other bird species, which was a primary objective of WCSP
program and Wildlife Services. Cummings et al. (1987) evaluated a similar program in the
early 1980s and found that bird use of decoy plots was largely influenced by the surrounding
landscape, but this included only blackbird species.
Here, we (1) compare bird use [excluding blackbirds which was described by Hagy et al.
(2008)], among WCSP, commercial sunflower, and other non-sunflower row-crop fields, (2)
describe the within-field and surrounding landscape characteristics of WCSP that were
related to bird use and (3) develop guidelines for future placement and management of
120 THE AMERICAN MIDLAND NATURALIST 164(1)
WCSP in the NGP so that agricultural producers, conservation planners and habitat
managers can collectively improve migratory bird habitat while retaining productive
agricultural practices.
METHODS
STUDY AREA
We surveyed 35 sites in late summer and fall 2004 and 2005 in the Prairie Pothole Region
(PPR) of east-central North Dakota. Each site included one WCSP, one nearby[#2.4 km
from WCSP] commercial sunflower field, and one nearby non-sunflower row-crop field.
Sites were located at least 4.5 km apart and the study area encompassed ,60,000 km
2
. Row-
crop agriculture (canola, flax, lentils, oats, soybeans, sunflower and wheat), grasslands,
shelterbelts and wetlands are the most common pre-migratory and stopover habitats
available to birds in this region (Stewart, 1975; USDA, 2008). Approximately 400,000 ha of
sunflower are planted annually in North Dakota (USDA, 2008), more than any other state.
Sunflower is an important commodity in this region due to local processing and refining
infrastructure.
FIELD METHODS
We surveyed 14 WCSP and paired fields from 24 Aug.–19 Oct. 2004 three times at2wk
intervals and 21 WCSP and paired fields from 10 Aug.–28 Oct. 2005 two times at4wk
intervals. We conducted fixed radius (50 m) bird point counts and vegetation
measurements at one site (1 WCSP [mean 58.0 ha, SE 50.1] and one nearby, randomly
selected commercial sunflower [mean 533.7 ha, SE 52.9] and non-sunflower row-crop field
[mean 526.9 ha, SE 52.3]) per day. We selected commercial fields where bird disturbance
and blackbird harassment effort from producers appeared to be minimal. We used aerial
photographs and ground surveys to measure field dimensions and divided all fields into
square 1-ha units prior to bird and vegetation surveys. We overlaid a square 1-ha grid pattern
over the aerial imagery of each field and selected locations for the bird point counts and
subsequent vegetation surveys. As field sizes varied (8–90 ha), we randomly selected 15%of
the total ha in each field, during each survey round, for bird point counts with a minimum
of 2 point counts per field. Given large field sizes, 15%was typically the maximum area one
observer could survey in one day. We chose a conservative count distance of 50 m as habitat
structure and consequently species detectability varied across field-types (Cunningham and
Johnson, 2006) and long-distance point counts can result in considerable error in
abundance estimates (Efford and Dawson, 2009). We did not count birds in adjacent ha to
reduce the chance of double-counting and used the same commercial sunflower and non-
sunflower row-crop fields during each survey round. We randomly selected the order of
fields for bird surveys each time we visited a site.
We used distance sampling methods where one observer estimated and recorded distance
to each bird seen flying or perching 0–50 m from the center of each ha beginning 30 min
after sunrise. We waited 3 min after arriving at the center of each survey ha and then
conducted an 8 min visual bird survey. A 2 m step-ladder was used in tall crops(e.g.,
sunflower and corn) to aid bird observations. One observer conducted all counts in 2004,
and two observers conducted all counts in 2005. In 2005, one observer completed all surveys
for a given site during a given survey round, individually and observers periodically practiced
bird identification and distance estimation together prior to point counts to increase
accuracy and precision of data collection. We did not conduct surveys in dense fog, when
raining or snowing, or when winds exceeded 20 kph (Ribic et al., 2009a).
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 121
Bird surveys were usually completed in all fields prior to 1000 (CDT). After completion,
we returned to each previous bird-survey ha and measured vegetation characteristics in two,
randomly selected 1-m
2
subplots per bird-survey ha. Measurements included crop density
(number of crop plants), vertical obstruction (measured with a 15 cm 32 m density board
placed in the middle of the subplot), row width, percent canopy cover (measured by placing
a spherical densitometer on the ground in the center of the subplot and counting
unobscured squares), weed density (number of non-crop plants), waste seed abundance
(seeds were vacuumed from the entire subplot using a portable vacuum), maximum plant
height and identification to family of all non-crop plant species present in the subplots.
Waste seed abundance was determined by washing all materials through a large (4.5 mm)
and small (0.3 mm) sieve, drying seeds and detritus, and manually removing, counting and
weighing seeds (natural and from the crop). Data from two subplots was averaged to
generate one set of vegetation metrics per survey ha before analysis. Sites were surveyed in
the same order within each round of surveys as the large size of the study area made
randomizing site visitation order logistically impractical.
In late summer 2004 and 2005, we visually surveyed and delineated all major land-uses
within 2.4 km of each WCSP (Cummings et al., 1987; Cunningham and Johnson, 2006). We
obtained aerial photography from the USDA-NRCS Geospatial Data Gateway (USDA, 2005)
and manually overlaid land use data within 2.4 km of each WCSP in ArcMap (ESRI, 2004).
Statistical analysis.—We counted blackbirds, but omitted them from analysis as they form
large foraging flocks and were periodically abundant enough to overwhelm analysis
compared to other species (Otis and Kilburn, 1988; Schaff, 2003). Hagy et al. (2008)
reported on blackbird use of WCSP and surrounding agricultural habitats. Other studies
have similarly omitted them from analysis due to their ubiquitous habitat use (Ribic et al.,
2009a) and behavior (Lokemoen and Beiser, 1997).
We used program DISTANCE (ver. 5.0; Thomas et al., 2005) to estimate bird densities
using crop- and species-specific detection functions adjusting observations for detectabil-
ities. We pooled observations among survey rounds, but not years or field types to estimate
detection functions. We used the multi-covariate distance sampling engine in Program
DISTANCE and detection functions for half-normal and hazard-rate key functions with
cosine, simple polynomial and hermite polynomial series expansions with stratification at
the site level (Marques et al., 2007). Different combinations of the above key functions and
series expansions allow variable fitting of the distance estimation curve, thus improving
accuracy of density estimates for each data set of bird observations due to different
detectabilities. We constructed all models with and without the covariates of plant height
and vertical cover, which were not strongly correlated. We selected the best models based on
Chi-square model fit statistics, Akaike’s information criterion (Akaike, 1974), shape and
structure of the detection probability and probability plots, and biological plausibility of the
density estimates (Buckland et al., 2001; Burnham and Anderson, 2002). We were not able to
reliably estimate species-specific densities for each site due to few or variable detections. We
grouped data into even intervals and left-truncated at 3 m for improved model fit after
examining distance histograms to detect any rounding of distances by observers (Buckland
et al., 2001).
We generated density estimates of all species combined (‘‘all birds’’) using global and
crop-specific detection functions and post stratification for all birds present in WCSP,
commercial sunflower, and non-sunflower row-crop fields (Alldredge et al., 2007).
Additionally, we estimated densities of species with at least 85 detections using species-
specific detection functions. We used a global detection function (generated by pooling
122 THE AMERICAN MIDLAND NATURALIST 164(1)
observations among all species) to generate species-specific density estimates for species
with fewer than 85 detections (Buckland et al., 2001, p. 240).
We used an information theoretic approach to model all bird density as a function of
habitat variables and constructed models using 22 habitat variables which included a priori-
selected within-field vegetation metrics and surrounding land-use variables using linear
mixed models (Proc MIXED in SAS, SAS Institute Inc., 2005). Models consisted of 2–5
independent variables based on our sample size (n 535) compared to the dependent
variable, all bird density.
We examined bird densities and all independent variables (Table 1) for correlations and
deviation from normality using Microsoft Excel Pop Tools and JMP (SAS Institute Inc.,
2005). Non-normal data were transformed using either ln(x +1) or sqrt(x) transformations
as appropriate to achieve a normal distribution and homogeneity of variance (Zar, 1999).
We designated year as a random effect and survey round as the repeated measure. We used
model selection based on Akaike’s second order Information Criterion (AIC
c
) to evaluate
all models (Akaike, 1974; Burnham and Anderson, 2002).
We formed 2 sets of models for WCSP, an a priori and post hoc model set. We constructed a
priori models based on previously published bird-habitat relationships and knowledge of
bird ecology in agricultural habitats before completing data collection and analysis (Best et
al., 1990, 1998, 2001; Crozier and Niemi, 2003). Post hoc models were formed because little
published information exists on fall bird use of agricultural crop fields and we desired to use
our experience gained while working in the study system to elucidate possible relationships
between birds and landscape factors. Furthermore, in exploratory studies where little
published information exists describing the study systems, it may be unreasonable to assume
that researchers can formulate all possible and biologically plausible models before data
collection (Hagy et al., 2008; Uyehara et al., 2008). Therefore, we formed post hoc models
using SAS Proc MIXED to regress each independent variable (Table 1) singularly with each
response variable and the random effect of year (null model), and ranked them from lowest
AICc value (the best performing independent variables) to highest (the worst performing
independent variable; Cunningham and Johnson, 2006; Hagy et al., 2008; Uyehara et al.,
2008). We used these AIC
c
scores and a posteriori knowledge of the ecological system to
construct only biologically plausible post hoc models (Anderson et al., 2000; Johnson and
Omland, 2004). The formation of only biologically plausible models has been recommend-
ed over running all possible models to avoid spurious findings (Johnson and Omland,
2004). We labeled all models by the variables they contained (intrinsic WCSP metrics,
extrinsic WCSP metrics and mixed metrics) to evaluate the relative influence of intrinsic
(habitat metrics measured from within each field such as crop density) versus extrinsic
[habitat characteristics measured outside of each field such as nearby (#2.4 km) wetland
ha] characteristics on bird density (Moore and Simons, 1992; Cunningham and Johnson,
2006). All final models were evaluated using SAS (Proc MIXED) and maximum likelihood
estimation as we varied the fixed effects throughout the final models in each set (Littell et
al., 1996; Riffell et al., 2006).
We assessed relative variable importance using only models in the 90%confidence set
whose weights collectively summed to $0.90, since several models were supported
(Burnham and Anderson, 2002; Riffell et al., 2006). We ranked all competing models by
DAIC
c
, calculated model weights and model averaged parameter estimates, and estimated
relative variable importance with variances from the 90%confidence set (Burnham and
Anderson, 2002). Interpretation of independent variable importance was based on model
averaged parameter estimates and the frequency of inclusion within the 90%confidence set.
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 123
TABLE 1.—Bird detections (N) and density estimates (birds/ha; D) with corresponding standard
errors (SE) and variation in density attributable to detection probability (%s
ˆ
2
) in WCSP, commercial
sunflower and non-sunflower row crops generated by fitting global detection probabilities in
DISTANCE to point-count observations
Species
WCSP SUNFLOWER CROP
NDSE%s
ˆ
2
NDSE%s
ˆ
2
NDSE%s
ˆ
2
American Goldfinch
Carduelis tristis 163 3.91 0.97 9.3 312 2.63 0.41 2 4 0.05 0.03 0.4
American Pipit
Anthus rubescens — — — — 40 0.33 0.24 0.1 — — — —
American Robin
Turdus migratorius 7 0.16 0.14 0.8 6 0.05 0.03 0.1 — — — —
Barn Swallow
Hirundo rustica 53 1.27 0.42 5.2 123 1.03 0.28 0.6 197 2.64 0.85 1.3
Blue Jay
Cyanocitta cristata — — — — 6 0.05 0.03 0.1 — — — —
Bobolink
Dolichonyx oryzivorus 16 0.38 0.23 1.6 31 0.26 0.15 0.1 — — — —
Brown Thrasher
Toxostoma rufrum 3 0.72 0.41 1.7 1 0.01 0.01 0 — — — —
Cedar Waxwing
Bombycilla cedrorum — — — — 2 0.01 0.01 0 — — — —
Chipping Sparrow
Spizella passerina 20 0.48 0.2 3.3 35 0.29 0.19 0.1 3 0.04 0.03 0.2
Clay-colored Sparrow
Spizella pallida 144 3.45 0.7 13.9 226 1.9 0.31 1.7 10 0.13 0.05 0.7
Common Yellowthroat
Geothlypis trichas 9 0.21 0.12 1.7 4 0.03 0.02 0.1 — — — —
Cooper’s Hawk
Accipiter cooperii 3 0.72 0.53 1 1 0.01 0.01 0 — — — —
Dark-eyed Junco
Junco hyemalis 16 0.38 0.29 1 24 0.2 0.13 0.1 — — — —
Dickcissel
Spiza americana — — — — 3 0.02 0.01 0.1 — — — —
Eastern Kingbird
Tyrannus tyrannus 3 0.07 0.07 0.6 — — — — 2 0.02 0.01 0.3
Evening Grosbeak
Coccothraustes vespertinus 1 0.02 0.02 0.6 — — — — — — — —
Field Sparrow
Spizella pusilla 6 0.14 0.1 1 2 0.01 0.01 0 5 0.06 0.05 0.2
124 THE AMERICAN MIDLAND NATURALIST 164(1)
Species
WCSP SUNFLOWER CROP
NDSE%s
ˆ
2
NDSE%s
ˆ
2
NDSE%s
ˆ
2
Fox Sparrow
Passerella iliaca — — — — 1 0.01 0.01 0 — — — —
Franklin’s Gull
Larus pipixcan — — — — — — — — 1 0.1 0.01 0.1
Gray Partridge
Perdix perdix 1 0.02 0.02 0.6 — — — — — — — —
Harris’s Sparrow
Zonotrichia querula 14 0.33 0.25 1 4 0.03 0.02 0.1 — — — —
Horned Lark
Eremophila alpestris — — — — — — — — 138 1.85 0.84 0.6
House Finch
Carpodacus mexicanus — — — — 6 0.05 0.05 0 — — — —
House Sparrow
Passer domesticus — — — — 1 0.01 0.01 0 — — — —
Killdeer
Charadrius vociferus 1 0.02 0.02 0.6 — — — — 10 0.13 0.08 0.4
Lapland Longspur
Calcarius lapponicus — — — — — — — — 12 0.16 0.13 0.2
Lark Sparrow
Chondestes grammacus — — — — — — — — 10 0.13 0.1 0.2
Lincoln’s Sparrow
Melospiza lincolnii 8 0.19 0.1 2 19 0.16 0.08 0.2 5 0.06 0.03 0.5
Marsh Wren
Cistothorus palustris 3 0.07 0.05 1 — — — — — — — —
Mourning Dove
Zenaida macroura 97 2.33 1.09 2.6 88 0.74 0.11 2 8 0.1 0.03 1.1
Northern Flicker
Colaptes auratus — — — — — — — — 1 0.01 0.01 0.1
Northern Harrier
Circus cyaneus — — — — 1 0.01 0.01 0 — — — —
Northern Waterthrush
Seiurus noveboracensis 1 0.02 0.02 0.6 — — — — — — — —
Orchard Oriole
Icterus spurius — — — — — — — — 2 0.02 0.02 0.1
Palm Warbler
Dendroica palmarum — — — — 1 0.01 0.01 0 — — — —
Ring-necked Pheasant
Phasianus colchicus 9 0.21 0.21 0.6 3 0.02 0.01 0.1 — — — —
TABLE 1.—Continued
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 125
Using bird densities generated by stratification by site in DISTANCE, we compared all
bird densities across all field types using repeated measures analysis of variance (ANOVA).
We assigned field type as the treatment effect, survey round as the repeated measure, and
year as a random effect in Proc MIXED (SAS Institute Inc., 2005). We used post-hoc Tukey
Species
WCSP SUNFLOWER CROP
NDSE%s
ˆ
2
NDSE%s
ˆ
2
NDSE%s
ˆ
2
Ruby-throated
Hummingbird
Archilochus colubris 1 0.02 0.02 0.6 1 0.01 0.01 0 1 0.01 0.01 0.1
Savannah Sparrow
Passerculus
sandwichensis 143 3.4 0.83 9.8 127 1.07 0.2 1.3 33 0.44 0.11 2
Sharp-tailed Grouse
Tympanuchus
phasianellus — — — — 25 0.21 0.15 0.1 1 0.01 0.01 0.1
Song Sparrow
Melospiza melodia 27 0.64 0.32 2.3 14 0.11 0.06 0.1 3 0.04 0.04 0.1
Swamp Sparrow
Melospiza geogiana 4 0.09 0.09 0.6 5 0.04 0.03 0.1 — — — —
Tennessee Warbler
Vermivora peregrina 1 0.02 0.02 0.6 — — — — — — — —
Tree Swallow
Tachycineta bicolor 31 0.74 0.34 2.7 49 0.41 0.17 0.3 74 0.99 0.35 1
Vesper Sparrow
Pooecetes gramineus 3 0.07 0.05 1 38 0.32 0.16 0.2 1 0.01 0.01 0.1
Western Meadowlark
Sturnella neglecta 10 0.24 0.18 1 17 0.14 0.08 0.2 6 0.08 0.03 0.8
White-crowned Sparrow
Zonotrichia leucophrys 6 0.14 0.1 1 1 0.01 0.01 0 — — — —
White-throated Sparrow
Zonotrichia albicollis 6 0.14 0.1 1.1 14 0.11 0.09 0.1 54 0.72 0.58 0.2
Yellow Warbler
Dendroica petechia 1 0.02 0.02 0.6 12 0.1 0.05 0.2 — — — —
Yellow-bellied Flycatcher
Empidonax flaviventris 2 0.04 0.04 0.6 5 0.04 0.04 0 1 0.01 0.01 0.1
Yellow-rumped Warbler
Dendroica coronata 2 0.04 0.04 0.6 51 0.43 0.19 0.2 — — — —
Unidentified Birds 204 218 95
Effective Detection
Radius 27.9 61.06 31.2 60.03 26.4 60.04
TABLE 1.—Continued
126 THE AMERICAN MIDLAND NATURALIST 164(1)
multiple comparisons (a50.05) to test for differences between bird densities in WCSP,
commercial sunflower, and non-sunflower row-crop fields (Zar, 1999). We used bird
densities and non-crop plant abundance to calculate Simpson’s Diversity Index for WCSP,
commercial sunflower and non-sunflower row-crop fields (McCune and Grace, 2002). We
similarly used a mixed model ANOVA to compare abundance of non-crop plants and
Simpson’s diversity among field types. We chose Simpson’s diversity index because of
concerns with other diversity indices expressed in Whittaker (1972).
RESULTS
Bird abundance.—We observed 34 bird species in WCSP (mean 54.2 species/ha), 37 in
commercial sunflower (mean 51.1 species/ha) and 24 in commercial non-sunflower row-
crops (mean 50.8 species/ha; Table 1). Birds occurred at greater densities (P ,0.01) but
similar diversities in WCSP (mean 524.4 birds/ha, SE 52.7; Div 50.88), followed by
commercial sunflower (mean 512.7 birds/ha, SE 51.7; Div 50.88) and non-sunflower row-
crops (mean 57.2 birds/ha, SE 51.1; Div 50.80; Fig. 1). We recorded sufficient detections
of individuals (n $85) to generate species-specific detection probabilities and densities for
four species in WCSP, five species in commercial sunflower and two species in other non-
sunflower row-crops (Table 2).
Influence of habitat variables.—The mean abundance of non-crop plants was similar (P 5
0.71) among WCSP (mean 553.9 forbs/m
2
,SE 511.9), commercial sunflower (mean 5
41.8 forbs/m
2
,SE 510.7) and other non-sunflower row-crops (mean 554.6 forbs/m
2
,SE 5
12.7). Most non-crop plants were in the Poaceae family in all three field types (61%in
WCSP, 57%in commercial sunflower, and 87%in non-sunflower row-crop fields); however,
Simpson’s diversity index for non-crop plant abundance varied (P ,0.01) among WCSP
(Div 50.96), commercial sunflower (Div 50.87) and non-sunflower row crops (Div 50.77).
FIG 1. Mean all bird abundance (2004–2005) in WCSP, commercial sunflower and other non-
sunflower row crops with associated standard errors.—Letters A, B and C represent a50.05 differences
(Tukey’s post hoc pair-wise comparisons among habitat types)
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 127
TABLE 2.—Bird abundance, species-specific detection probabilities (P), effective detection radius (EDR) and density estimates (birds/ha; D) with
coefficients of variation and 95%confidence intervals of species with $85 detections. The best-fitting model is described by detection key function, series
expansion and covariates
Field Species N P EDR D %CV 95%CI Model
WCSP American Goldfinch 163 0.24 28.47 3.78 39.13 1.8–7.96 Hazard Rate +Cosine
Carduelis tristis
Clay-colored Sparrow 144 0.28 30.17 2.92 23.58 1.88–4.71 Hazard Rate +Hermite Polynomial
Spizella pallida
Mourning Dove 97 0.14 22.06 3.75 53.23 1.39–10.09 Hazard Rate +Hermite Polynomial
Zenaida macroura
Savannah Sparrow 143 0.102 18.27 8.01 22.12 4.64–13.84 Hazard Rate +Simple Polynomial +Tall
Passerculus sandwichensis
Sunflower American Goldfinch 312 0.28 30.4 2.76 20.64 1.84–4.13 Hazard Rate +Hermite Polynomial
Carduelis tristis
Barn Swallow 123 0.1 18.2 3.03 43.4 1.3–6.8 Hazard Rate +Hermite Polynomial
Hirundo rustica
Clay-colored Sparrow 226 0.26 29.17 2.18 27.7 1.27–3.75 Half Normal +Cosine
Spizella pallida
Mourning Dove 88 0.22 27.26 0.97 48.12 0.39–2.40 Hazard Rate +Simple Polynomial
Zenaida macroura
Savannah Sparrow 127 0.35 33.88 0.9 27.91 0.52–1.56 Hazard Rate +Simple Polynomial
Passerculus sandwichensis
Crop Barn Swallow 197 0.14 21.5 4.04 35.52 2.04–8.02 Hazard Rate +Simple Polynomial
Hirundo rustica
Horned Lark 138 0.38 35.4 1.02 48.01 0.41–2.54 Hazard Rate +Simple Polynomial
Eremophila alpestris
* Key Functions (Half Normal, Hazard Rate), Series Expansions (Hermite Polynomial, Simple Polynomial, Cosine), Covariate describing plant height
(Tall)
128 THE AMERICAN MIDLAND NATURALIST 164(1)
Total non-crop seed weight was similar (P 50.46) among WCSP (mean 53.2 kg/ha, SE 5
0.8), commercial sunflower (mean 53.6 kg/ha, SE 50.9) and other non-sunflower row-
crops (mean 52.1 kg/ha, SE 50.7). We did not compare other vegetation metrics (i.e., row
width, vertical obstruction, canopy coverage, etc.) among the three crop types as they were
dependent on the crop type itself and varied considerable, especially in non-sunflower row-
crop fields [canola (n 52), flax (n 51), lentils (n 52), soybeans (n 518) and wheat (n 5
12)].
We evaluated 37 models constructed both a priori (n 520) and post hoc (n 517)
describing our dependent variable (all bird density) and a variety of explanatory variables in
WCSP. The 90%confidence set of best models (summed model weight 50.93, DAIC
c
50–
5.7; Table 3) included 12 models. Seven of these 12 top models included only within-field
(intrinsic) habitat metrics and all 12 contained at least one intrinsic metric (Moore and
Simons, 1992). The top model (w
i
50.26) for birds in WCSP was formulated post hoc and
included the intrinsic variables crop plant density (+), abundance of Brassicaceae plants (+),
and maximum plant height (+; Table 3). No models consisting of only extrinsic habitat
metrics (landscape variables) or models formed a priori were included in the 90%
confidence set.
Crop density (w
i
51.00) and maximum crop plant height (w
i
50.68) were positively
associated with all bird density and present in 12 and seven top models, respectively. Non-
crop plants in the families Chenopodiaceae [w
i
50.37; fireweed (Kochia spp.) and goosefoot
(Chenopodium spp.)] and Brassicaceae [w
i
50.88; wild mustards (Sinapsis sp. and Brassica
spp.)] were positively associated with all bird density in four and seven top models,
respectively; however, Polygonaceae [w
i
50.15; buckwheat (Polygonum spp.)] was negatively
associated with all bird density in one top model (Table 4).
TABLE 3.—Bird 90%confidence set with formulation technique [post hoc or a priori and intrinsic
(within-field), extrinsic (surrounding land use) or mixed metrics], log-likelihood (£), and DAIC
c
values,
model weight (w
i
) and evidence ratios (ER)
Model Variables £ DAIC
c
w
i
ER
post hoc—intrinsic Brassicaceae, crop density, crop height 206.7 0.0 0.26 1.00
post hoc—mixed Brassicaceae, Chenopodiaceae, crop
density, crop height, Polygonaceae
202.7 1.2 0.15 1.82
post hoc—mixed Brassicaceae, Chenopodiaceae, crop
density, crop height
205.8 1.7 0.11 2.34
post hoc—mixed Brassicaceae, crop density, wetlands 208.5 1.8 0.11 2.46
post hoc—intrinsic Brassicaceae, crop density 211.0 1.9 0.10 2.59
post hoc—intrinsic Brassicaceae, Chenopodiaceae, crop density 209.9 3.2 0.05 4.95
post hoc—intrinsic crop density, crop height 212.9 3.8 0.04 6.69
post hoc—intrinsic Brassicaceae, Chenopodiaceae,
crop density, crop height, weeds
206.6 4.1 0.03 7.77
post hoc—mixed crop density, crop height, sunflowers 211.6 5.0 0.02 12.18
post hoc—mixed crop density, sunflowers 214.6 5.5 0.02 15.64
post hoc—mixed crop density, grasslands 214.9 5.7 0.02 17.29
post hoc—mixed crop density, crop height, grasslands 212.4 5.7 0.02 17.29
* Brassicaceae—abundance of plants from this family, Chenopodiaceae—abundance of plants from
this family, Polygonaceae—abundance of plants from this family, crop density—mean crop plants/m
2
,
crop height—mean maximum crop height, sunflowers—area of surrounding sunflower, grasslands—
area of surrounding grass, weeds—abundance of non-crop plants, wetlands—area of surrounding
wetlands
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 129
DISCUSSION
All bird and most individual species densities were greatest in WCSP compared to other
sunflower and non-sunflower crop fields. Bird densities in WCSP were more strongly
associated with habitat variables within-WCSP than surrounding land use; however, forb and
non-crop seed abundance was similar among all field-types. We noted that in most non-
sunflower crop fields, non-crop plants were small and often immature, although we did not
quantify physical characteristics of non-crop plants in fields. In comparison, non-crop plants
were often large and mature, flowering or with seed, in sunflower fields. This is likely due to
later harvest and reduced use of herbicides in sunflower fields compared to other crops,
especially soybeans. Plants of the family Brassicaceae continue to bloom in late summer and
fall possible attracting insects and thus insectivores. Kochia spp. and other Chenopodiaceae
plants prolifically produce seeds that are eaten by granivorous birds. Furthermore, the
smaller size of WCSP, resulting in greater edge to interior ratio compared to other fields,
may have also contributed to higher bird densities (Yahner, 1988).
Previous studies have shown that at least 94 bird species use sunflower and other crop
fields during the fall and spring in the NGP (Schaaf, 2003; Galle, 2005; Hagy et al., 2007).
Hagy et al. (2007) reported the occurrence of 12 species of conservation concern in the PPR
using sunflower during migration (USDI, 2002; Rich et al., 2004; Butcher et al., 2007). We
observed one species of conservation concern (USDI, 2002; Rich et al., 2004; Butcher et al.,
2007) in WCSP (bobolink, Dolichonyx oryzivorus,n516), two in commercial sunflower
(bobolink, n 531; northern harrier, Circus cyaneus,n51) and two in non-sunflower crop
fields (Franklin’s gull, Larus pipixcan,n51; lark sparrow, Chondestes grammacus,n510)
during point counts. Compared to other row crops, sunflower was more commonly used by
migratory birds in the fall and spring in the NGP (Galle, 2005; Hagy et al., 2007).
Surprisingly few studies have examined bird use of active agricultural fields, given
widespread cultivation of lands in North America. Best et al. (2001) examined the influence
of landscape composition on bird use in row-crop fields, but did not compare that with
intrinsic characteristics of the crop fields themselves. Lokemoen and Beiser (1997) found
that bird density was greatest in the fall in conventional cropland fields compared to
minimum tillage and organic–management crop fields. We have noted recent works that
excluded agriculture lands from habitat analyses (Ribic et al., 2009a) even though
agriculture dominated the study area and birds used those habitats (Best et al., 2001; Hagy
et al., 2007), and others that refer to agricultural lands as ‘‘hostile’’ (Ribic et al., 2009b).
TABLE 4.—Variables included in the 90%confidence set of candidate models with the number of
models entered, summed model weights (Pv) and model-averaged parameter estimate ( ~
b) with
standard error (SE)
Variable Models Pvi~
bSE
Crop density 12 1.000 0.382 0.903
Crop height 7 0.682 0.606 1.114
Brassicaceae 7 0.883 0.194 0.449
Chenopodiaceae 4 0.372 0.030 0.103
Sunflowers 2 0.042 20.002 0.009
Grasslands 2 0.033 0.005 0.015
Wetlands 1 0.116 20.022 0.047
Weeds 1 0.037 20.002 0.011
Polygonaceae 1 0.156 20.018 0.045
130 THE AMERICAN MIDLAND NATURALIST 164(1)
While agricultural lands may exhibit considerable temporal structural variability compared
to grassland habitats, we note the variety of species that use them during fall in North
Dakota. Furthermore, we observed greater bird densities in sunflower and especially in
WCSP than previously reported during fall in fallow, sunflower and wheat fields (Lokemoen
and Beiser, 1997); in cornfields during the breeding season (Best et al., 1990); in Great
Plains forest fragments in spring (Martin, 1980); and in Conservation Reserve Program
grasslands and row-crop fields in winter (Best et al., 1998).
Birds were more closely associated with vegetation characteristics within WCSP than
landuse outside of the lure plots. Seven of the 12 top models contained only within-field
habitat metrics and the top four model averaged scores were from within-field metrics.
Contrary to results describing response of blackbirds to surrounding landscape variables
reported by Hagy et al. (2008) and Cummings et al. (1987), non- blackbird avian species may
select crop fields using proximate cues within-fields rather than landscape composition at
the 2.4 km scale in our study area. In east-central North Dakota, abundant wetlands,
grasslands and agricultural fields may facilitate a selection response to the region (in this
case, our ,60,000 km
2
study area) rather than small subunits of the landscape, such as our
2.4 km site area. It is possible that birds in this study selected the landscape at a scale larger
than 2.4 km, but due to logistical and temporal restraints we were not able to test this
hypothesis.
We determined that tall ($1.3 m), densely planted within rows ($5 crop plants /
linear m), WCSP with some non-crop plants (forbs within Brassicaceae and Chenopo-
diaceae families) resulted in the greatest bird densities in WCSP. Although landscape
variables received less model weight, ha of surrounding grasslands were positively
associated while ha of sunflowers and wetlands were negatively associated with all bird
density in WCSP. Commercial sunflower and wetlands with emergent vegetation likely
provide alternative habitat for migrating birds, which may explain a negative correlation
with bird densities in WCSP. Additionally, blackbirds greatly reduced sunflower seed
abundance in some WCSP near wetlands, which may have reduced food resourcesfor
other granivorous birds (Hagy et al., 2008). We recommend planting WCSP with dense
plant spacing within rows, and using management practices that allow non-crop plants to
persist. Wide row spacing and initial site preparation with little subsequent between-row
tillage will likely result in tall, vegetatively diverse WCSP attractive to a variety of bird
species. Furthermore, planting WCSP near cattail-dominated wetlands and shelterbelts
that blackbirds use as roosting and loafing sites may increase effectiveness as a pest-
management system (Hagy et al.,2008).
Our results are consistent with the findings of Cunningham and Johnson (2006) that
landscape variables alone were not as closely associated with bird distributions as either
models consisting of only within-field (intrinsic proximate cues) or within-field and
landscape-level variables. These authors also reported closer bird-landscape associations at
large scales (800–1600 m) for most grassland species they encountered; however, at the
2.4 km scale, we did not detect similar associations. Similarly, Ribic et al. (2009a) found that
grassland birds were more strongly associated with habitat type than surrounding landscape
metrics. We suggest further research be conducted at large (.2.4 km) or multiple spatial
and temporal scales in agricultural fields during non-breeding periods to examine both
intrinsic and extrinsic habitat variables in comparison to bird densities.
Hagy et al. (2008) reported the cost of planting 41 WCSP in 2004 and 2005 for blackbird
damage prevention to exceed direct savings from damage amelioration by a factor of 2.3:1.
However, individual producers were not subject to a deficit because of the cost-share
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 131
program. Furthermore, benefits of migratory bird and wildlife habitat could further offset
costs substantially to sunflower producers. Cost-share planting, land-lease for hunting, and
difficult to quantify factors such as migratory bird habitat likely increase the value of WCSP
to producers.
It is difficult to quantify non-market values of wildlife habitat, especially for non-game
animals (Boyer and Polasky, 2004). However, we approximated the value of wildlife habitat
of WCSP using economic incentives offered by the Private Land Initiative Food Plot
Program for sportsman implemented by the North Dakota Game and Fish Department. The
food plot program awards as much as $US46/ha to private landowners for planting rowcrop
food plots for wildlife and allowing walk-in access. Combining these economic incentives, as
a surrogate valuation of wildlife food plots, with blackbird damage reduction would result in
a 1.8:1 cost (subsidization $US375/ha) to benefits (average value of damage reduction
provided by plots +value of food plots for wildlife) ratio on average. To implement the
program at a 1:1 cost-benefit ratio and make subsidization economically sustainable for
Wildlife Services, other ecological services and non-market value of WCSP would need to
total $US195/ha.
Wildlife Services subsidized sunflower producers $US375/ha to plant WCSP in 2004–
2005. A highly competitive applicant pool of sunflower producers indicated that
participation in the program was economically beneficial for producers at that rate. If
producers were willing to participate in the program for reduced compensation, then it is
likely that the benefits might further outweigh the costs associated with the WCSP program.
Additionally, these cost-benefit analyses was applied using all of the plots in our study area
and mean blackbird damage estimates from Hagy et al. (2008). Some individual WCSP were
economically viable considering damage amelioration alone because they were depredated
upwards of 80%by blackbirds, yet still provided fall-winter habitat and foods for wildlife.
Using placement and management recommendations from this study and Hagy et al.
(2008), WCSP can be maximized to provide an economically viable wildlife habitat/non-
lethal blackbird damage control system. We suggest further implementation of the WCSP
program using our management recommendations and further evaluation including
economic quantification of benefits provided by WCSP.
WCSP can accommodate a wide variety of migratory birds while providing a row crop/
blackbird damage control system for farmers in the northern Great Plains. Integrative pest
management/wildlife management systems may become increasingly important in the NGP
as sunflower acreages have declined by almost 50%while corn (+250%) and soybeans
(+100%), both with glyphosate tolerant varieties, have become increasingly common in
North Dakota since 1998 (USDA, 2008). Conversion of sunflower and grassland habitats in
the NGP to other habitats could negatively affect migratory birds given their abundance in
these habitats compared to other croplands (Otis and Kilburn, 1988). Furthermore,
decreasing sunflower could concentrate blackbirds and require more intensive and lethal
pest management techniques, which may negatively affect other wildlife. If economically
viable, continued sunflower production and WCSP implementation may benefit migratory
birds by providing better habitat than other row-crops in the NGP.
Acknowledgments.—We thank the United States Department of Agriculture Wildlife Service’s National
Wildlife Research Center, North Dakota Wildlife Services and North Dakota State University
Department of Biological Sciences for funding and support. We thank NDSU graduate students and
field and laboratory assistants for their support and dedicated service. We thank S. K. Riffell for
suggestions pertaining to our statistical analysis and R. M. Bush, G. M. Forcey, E. H. Hagy, M. McConnell
and M. L. Schummer for evaluating early versions of this manuscript.
132 THE AMERICAN MIDLAND NATURALIST 164(1)
LITERATURE CITED
AKAIKE, H. 1974. A new look at the statistical model identification. IEEE Trans. Auto. Contr.,19:716–723.
ALLDREDGE, M. W., K. H. POLLOCK,T.R.SIMONS AND S. A. SHRINER. 2007. Multiple-species analysis of point
count data: a more parsimonious modeling framework. J. Appl. Ecol.,44:281–290.
ANDERSON, D. R., K. P. BURNHAM AND W. L. THOMPSON. 2000. Null hypothesis testing: problems,
prevalence, and an alternative. J. Wildl. Manage.,64:912–923.
BERGIN, T. M. 1992. Habitat selection by the western kingbird in western Nebraska: a hierarchical
analysis. Condor,94:903–911.
BEST, L. B., T. M. BERGIN AND K. E. FREEMARK. 2001. Influence of landscape composition on bird use of
rowcrop fields. J. Wildl. Manage.,65:442–449.
———, R. C. WHITMORE AND G. M. BOOTH. 1990. Use of corn fields by birds during the breeding season:
the importance of edge habitat. Am. Midl. Nat.,123:84–99.
———, H. CAMPA,K.E.KEMP,R.J.ROBEL,M.R.RYAN,J.A.SAVIDGE,H.P.WEEKS,JR.AND S. R. WINTERSTEIN.
1998. Avian abundance in CRP and crop fields during winter in the midwest. Am. Midl. Nat.,
139:311–324.
BOYER,T.AND S. POLASKY. 2004. Valuing urban wetlands: a review of non-market valuation studies.
Wetlands,124:744–755.
BUCKLAND, S. T., D. R. ANDERSON,K.P.BURNHAM,J.L.LOCKE,D.L.BORCHERS AND L. THOMAS. 2001.
Introduction to distance sampling: estimating abundance of biological populations. Oxford
University Press, Inc., New York, New York. 432 p.
BURNHAM,K.P.AND D. R. ANDERSON. 2002. Model selection and multimodel inference. Spring Verlag,
Inc., New York, New York. 488 p.
BUTCHER, G. S., D. K. NIVEN,A.O.PANJABI,D.N.PASHLEY AND K. V. ROSENBURG. 2007. The 2007 watchlist for
United States birds. Am. Birds,61:18–25.
CHARLET, L. D., G. J. BREWER AND B. A. FRANZMAN. 1997. Sunflower insects, p. 183–261. In: A. A. Schneiter
(ed.). Sunflower, Technology, and Production. Agronomy Monograph, No. 35. American
Society of Agronomy, Crop Science Society of America, and Soil Science Society of America,
Madison, Wisconsin.
CROZIER,G.E.AND G. J. NIEMI. 2003. Using patch and landscape variables to model abundance in a
naturally heterogeneous landscape. Can. J. Zool.,81:441–452.
CUMMINGS, J. L., J. L. GUARINO,C.E.KNITTLE AND W. C. ROYAL,JR. 1987. Decoy plantings for reducing
blackbird damage to nearby commercial sunflower fields. Crop Prot.,6:56–60.
CUNNINGHAM,M.A.AND D. A. JOHNSON. 2006. Proximate and landscape factors influence grassland bird
distributions. Ecol. Appli.,16:1062–10-75.
EFFORD,M.G.AND D. K. DAWSON. 2009. Effect of distance-related heterogeneity on population size
estimates from point counts. Auk,126:100–111.
ENVIRONMENTAL SYSTEMS RESEARCH INSTITUTE,INC. (ESRI). 2004. ArcMap version 9.1. Redlands, California.
FLATHER,C.H.AND J. R. SAUER. 1996. Using Landscape Ecology to test Hypotheses about large-scale
abundance patterns in migratory birds. Ecology,77:28–35.
GALLE, A. G. 2005. Avian use of harvested crop fields during spring migration through the southern drift
plains regions of North Dakota. M.S. thesis. North Dakota State University, Fargo. 146 p.
HAGY, H. M., G. M. LINZ AND W. J. BLEIER. 2007. Are sunflower fields for the birds? p. 61–71. In: D. L.
Nolte, W. M. Arjo and D. L. Stalman (eds.). Proceedings of the Twelfth Wildlife Damage
Management Conference. Corpus Christie, Texas.
———, ——— AND ———. 2008. Optimizing decoy crops for blackbird control in commercial sun-
flower. Crop Prot.,27:1442–1447.
JOHNSON, D. H. 1980. The comparison of usage and availability measurements for evaluating resource
preference. Ecology,61:65–71.
JOHNSON,J.B.AND K. S. OMLAND. 2004. Model selection in ecology and evolution. Trends Ecol. Evol.,
19:101–108.
KLEINGARTNER, L. 2003. Sunflower losses to blackbirds: an economic burden, p. 13–14. In: G. M. Linz
(ed.). Management of North American blackbirds. National Wildlife Research Center, Fort
Collins, Colorado, USA.
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 133
KOFORD,R.R.AND L. B. BEST. 1996. Management of agricultural landscapes for the conservation of
neotropical migratory birds, p. 68–88. In: F. Thompson (ed.). Management of midwestern
landscapes for the conservation of neotropical migratory birds. USDA, Forest Service General
Technical Report, NC-187. North Central Forest Experiment Station, St. Paul, Minnesota.
KOLASA, J. 1989. Ecological systems in hierarchical perspective: breaks in community structure and other
consequences. Ecology,70:36–47.
KOTLIAR,N.B.AND J. A. WIENS. 1990. Multiple scales of patchiness and patch structure: a hierarchical
framework for the study of heterogeneity. Oikos,59:253–260.
LINZ, G. M., H. J. HOMAN,L.B.PENRY AND P. MASTRANGELO. 2003. Reducing blackbird–human conflicts in
agriculture and feedlots: new methods for an integrated management approach, p. 21–24. In:
G. M. Linz (ed.). Management of North American blackbirds. National Wildlife Research
Center, Fort Collins, Colorado.
———, D. A. SCHAAF,P.MASTRANGELO,H.J.HOMAN,L.B.PERRY AND W. J. BLEIER. 2004. Wildlife
conservation sunflower plots as a dual purpose wildlife management strategy, p. 291–294. In:
R. M. Timm and W. B. Gorenzel (eds.). Proceedings of the 21st Vertebrate Pest Conference.
Visalia, California.
LITTELL, R. C., G. A. MILLIKEN,W.W.STROUP ,R.D.WOLFINGER AND O. SCHABENBERGER. 1996. SAS system for
mixed models. SAS Institute, Cary, North Carolina. 814 p.
LOKEMOEN,J.T.AND J. A. BEISER. 1997. Bird use and nesting in conventional, minimum tillage, and
organic cropland. J. Wildl. Manage.,61:644–655.
LUBOWSKI, B. N., M. VESTERBY,S.BUCHOLTZ,A.BAEZ AND M. J. ROBERTS. 2006. U. S. Department of
Agriculture Economics Research Service - major uses of land in the United States, 2002. EIB.
14. 54 p. http://www.ers.usda.gov/Data/MajorLandUses/MLUsummary tables.pdf.
MARQUES, T. A., L. THOMAS,S.G.FANCY AND S. T. BUCKLAND. 2007. Improving estimates of bird density
using multiple-covariate distance sampling. Auk,124:1229–1243.
MARTIN, T. E. 1980. Diversity and abundance of spring migratory birds using habitat islands on the Great
Plains. Condor,82:430–439.
MCCUNE,B.AND J. B. GRACE. 2002. Analysis of ecological communities. MjM Software Design, Gleneden
Beach, Oregon. 300 p.
MOORE, F. R., S. A. GAUTHREAUX,JR., P. KERLINGER AND T. R. SIMONS. 1995. Habitat requirements during
migration: important link in conservation, p. 121–144. In: T. Martin and D. M. Finch (eds.).
Ecology and management of neotropical migratory birds. Oxford University Press, New York,
New York.
——— AND D. A. ABORN. 2000. Mechanisms of en route habitat selection: how do migrants make habitat
decisions during stopover. Studies Avian Bio.,20:34–42.
——— AND T. R. SIMONS. 1992. Habitat suitability and stopover ecology of neotropical landbird migrants,
p. 345–355. In: J. M. Hagan and D. W. Johnson (eds.). Ecology and conservation of neotropical
migrant landbirds. Smithsonian Institution Press, Washington D.C.
MURPHY, M. T. 2003. Avian population trends within the evolving agricultural landscape of eastern and
central United States. Auk,120:20–34.
OTIS,D.L.AND C. M. KILBURN. 1988. Influence of environmental factors on blackbird damage to
sunflower. Fish and Wildlife Service Technical Report 16. U.S. Department of the Interior, Fish
and Wildlife Service, Bethesda, Maryland. 11 p.
PEARSE, A. T. 2007. Design, evaluation, and applications of an aerial survey to estimate abundance of
wintering waterfowl in Mississippi. Dissertation, Mississippi State University, Mississippi State,
USA. 170 p.
PEER, B. D., H. J. HOMAN,G.M.LINZ AND W. J. BLEIER. 2003. Impact of Blackbird damage to sunflower:
bioenergetic and economic models. Ecol. Appl.,13:248–256.
PETERJOHN, B. P. 2003. Agricultural landscapes: can they support healthy bird populations as well as farm
products? Auk,120:14–19.
RIBIC, C. A., M. J. GUZY AND D. W. SAMPLE. 2009a. Grassland bird use of remnant prairie and conservation
reserve program fields in an agricultural landscape in Wisconsin. Am. Midl. Nat.,161:110–122.
134 THE AMERICAN MIDLAND NATURALIST 164(1)
———, R. R. KOFORD,J.R.HERKERT,D.H.JOHNSON,N.D.NIEMITH,D.E.NAUGLE,K.K.BAKKER,D.W.
SAMPLE AND R. B. RENFREW. 2009b. Area sensitivity in North American grassland birds: patterns
and processes. Auk,126:233–244.
RICH, T. D., C. J. BEARDMORE,H.BERLANGA,P.J.BLANCHER,M.S.W.BRADSTREET,G.S.BUTCHER,D.W.
DEMAREST,E.H.DUNN,W.C.HUNTER,E.E.IN
˜IGO-ELIAS,J.A.KENNEDY,A.M.MARTELL,A.O.
PANJABI,D.N.PASHLEY,K.V.ROSENBERG,C.M.RUSTAY,J.S.WENDT AND T. C. WILL. 2004. Partners
in flight North American landbird conservation plan. Cornell Laboratory of Ornithology,
Ithaca, New York. http://www.partnersinflight.org/cont_ plan/PIF4_ AppendicesWEB.pdf
RIFFELL, S. K., T. BURTON AND M. MURPHY. 2006. Birds in depressional forested wetlands: area and habitat
requirements and model uncertainty. Wetlands,26:107–118.
SAS INSTITUTE INC., SAS 9.1.3. 2000–2005. The mixed procedure, Cary, North Carolina, USA.
SAUER, J. R., J. E. HINES AND J. FALLON. 2008. The North American Breeding Bird Survey, Results and
Analysis 1966–2007. Version 5.15.2008. USGS Patuxent Wildlife Research Center, Laurel,
Maryland, USA.
SCHAAF, D. A. 2003. Avian use of ripening sunflower fields. M.S. Thesis, North Dakota State University,
Fargo, North Dakota. 113 p.
STEWART, R. E. 1975. Breeding birds of North Dakota. Tri-college center for environmental studies.
Fargo, North Dakota. 295 p.
STONE,C.P.AND C. R. DANNER. 1980. Autumn flocking of red-winged blackbirds in relation to
agricultural variables. Am. Midl. Nat.,130:196–199.
THOMAS, J. L., S. STRINDBERG,F.F.C.MARQUES,S.T.BUCKLAND,D.L.BORCHERS,D.R.ANDERSON,K.P.
BURNHAM,S.L.HEDLEY,J.H.POLLARD,J.R.B.BISHOP AND T. A. MARQUES. 2005. DISTANCE
version 5.0 beta 5. Research Unit for Wildlife Population Assessment, University of St. Andrews,
United Kingdom.
U. S. DEPARTMENT OF AGRICULTURE [USDA]. 2008. Statistics by state – Quick stats. National Agricultural
Statistics Service, Fairfax, Virginia. http://www.nass.usda.gov/Statistics_by_State/North_
Dakota/index.asp.
U.S. DEPARTMENT OF AGRICULTURE NATIONAL RESOURCES CONSERVATION SERVICE [USDA]. 2005. Geospatial
data gateway. http://datagateway.nrcs. usda.gov.
U. S. DEPARTMENT OF INTERIOR [USDI]. 2002. Birds of conservation concern 2002. USDI, Fish and Wildlife
Service, Division of Migratory Bird Management, Arlington, Virginia. http://www.migratorybirds.
fws.gov/reports/bcc2002.pdf.
UYEHARA, K. J., A. ENGILIS,JR.AND B. D. DUGGER. 2008. Wetland features that influence occupancy by the
endangered Hawaiian duck. Wilson J. Ornith.,120:311–319.
WHITTAKER, R. H. 1972. Evolution and measurement of species diversity. Taxon,21:213–251.
YAHNER, R. H. 1988. Changes in wildlife communities near edges. Conserv. Biol.,2:333–339.
ZAR, J. H. 1999. Biostatistical analysis, 3rd ed. Prentice Hall, Englewood Cliffs, New Jersey. 944 p.
SUBMITTED 3JUNE 2009 ACCEPTED 11 NOVEMBER 2009
2010 HAGY ET AL.: SUNFLOWER CROPS AS BIRD HABITAT 135
