Effect of Predator Control on Reproductive Success and Hen Survival of Attwater's Prairie-chicken

Abstract

From 1980-1981, we tested the hypothesis that removal of potential nest predators would increase the reproductive success of the endangered Attwater's prairie-chicken (Tympanuchus cupido attwateri). Striped skunks (Mephitis mephitis, N = 74), opossums (Didelphis virginiana, N = 83), and raccoons (Procyon lotor, N = 9) were removed from a 522-ha predator removal area (PR) during February-June 1980 and 1981. Predator indices were lower (P < 0.002) and prairie-chicken nest success was higher (82% vs. 33%, P < 0.019) in the PR than a 620-ha control area (CO). Breeding season hen survival was <9% on both areas and survival curves were different between PR and CO (P < 0.015). Small sample size caused by declining populations and treatment effects that were compounded with site effects make our results equivocal. Managers may need to consider predator man-agement of a diverse group of species that prey on prairie-chicken adults and nests for a control program to be effective. The Attwater's prairie-chicken is an endangered grouse inhabiting areas of the Gulf Coastal Prairie in Texas. Prairie-chickens are a ground-nesting species and many nests fail to hatch because of destruction by predators or other causes (Lehmann 1941, Horkel et al. 1978, Lutz 1979, and Lawrence 1982). Several authors have recommended predator management to increase productivity of ground-nesting birds (Stoddard and Komarek 1941, Anderson 1957, Livezey 1981, Sargeant and Arnold 1984, Greenwood 1986). Other studies have docu-mented positive effects of predator control on nesting success (Balser et al. 1968, Chesness et al. 1968, Trautman et al. 1974).

Full text

Effect of Predator Control on Reproductive
Success and Hen Survival of Attwater's
Prairie-chicken
Jeffrey S. Lawrence,1 Department of Wildlife and Fisheries
Sciences,
Texas
A&M
University,
College Station, TX 77843
Nova J. Silvy, Department of Wildlife and Fisheries Sciences, Texas
A&M
University,
College Station, TX 77843
Abstract: From 1980-1981, we tested the hypothesis that removal of potential nest
predators would increase the reproductive success of the endangered Attwater's
prairie-chicken (Tympanuchus cupido attwateri). Striped skunks (Mephitis mephitis,
N = 74), opossums (Didelphis virginiana, N = 83), and raccoons (Procyon lotor,
N = 9) were removed from a 522-ha predator removal area (PR) during February-
June 1980 and 1981. Predator indices were lower (P < 0.002) and prairie-chicken
nest success was higher (82% vs. 33%, P < 0.019) in the PR than a 620-ha control
area (CO). Breeding season hen survival was <9% on both areas and survival
curves were different between PR and CO (P < 0.015). Small sample size caused
by declining populations and treatment effects that were compounded with site
effects make our results equivocal. Managers may need to consider predator man-
agement of a diverse group of species that prey on prairie-chicken adults and nests
for a control program to be effective.
Proc. Annu. Conf. Southeast. Assoc. Fish and Wildl. Agencies 49:275-282
The Attwater's prairie-chicken is an endangered grouse inhabiting areas of
the Gulf Coastal Prairie in Texas. Prairie-chickens are a ground-nesting species
and many nests fail to hatch because of destruction by predators or other causes
(Lehmann 1941, Horkel et al. 1978, Lutz 1979, and Lawrence 1982). Several
authors have recommended predator management to increase productivity of
ground-nesting birds (Stoddard and Komarek 1941, Anderson 1957, Livezey
1981,
Sargeant and Arnold 1984, Greenwood 1986). Other studies have docu-
mented positive effects of predator control on nesting success (Balser et
al.
1968,
Chesness et al. 1968, Trautman et al. 1974).
1 Present address: Wetland Wildlife Populations and Research Group, Minnesota Department
of Natural Resources, 102 23rd Street, Bemidji, Minnesota
56601.
1995 Proc. Annu.
Conf.
SEAFWA

276 Lawrence and Silvy
Attwater's prairie-chickens declined from an estimated 1,584 in 1980 (Law-
rence and Silvy 1980), the first year of this study, to 68 in 1995 (U.S. Fish and
Wildl. Serv., unpubl. data). Remaining populations of Attwater's prairie-
chickens are "island populations" (Lawrence and Silvy 1980). Excessive nest
losses can occur on such units because predators also are attracted to localized
areas (Braun et al. 1978, Sargeant and Arnold 1984). While habitat loss and
degradation are the major reasons for the endangered status of Attwater's
prairie-chicken (Lehmann 1968), strategies for maintaining viable populations
on managed areas need to be investigated. Our objective was to determine if
nest success and productivity of Attwater's prairie-chickens could be increased
by controlling major nest predators.
We are indebted to owners and staff of
T.
O'Connor Brothers' River Ranch
for access, lodging, and support. We appreciate the support of several individu-
als who assisted in the field and those who reviewed earlier drafts of the manu-
script. The Caesar Kleberg Research Program in Wildlife Ecology at Texas
A&M University provided funding. This is Contribution No. 21262, Texas Ag-
ricultural Experiment Station.
Study Areas
Unreplicated treatments were in the 6,400-ha Lake Pasture of the O'Con-
nor Brothers' River Ranch, about 28.8 km northeast of Refugio, Texas. Moder-
ate,
continuous grazing of
1
animal unit (=1 cow calf)/6.5 ha was maintained
throughout the study. Elevation of the study areas varied from 15.2 to 17.6 m.
Lake Pasture was intersected by 2 man-made drainages with intermittent flow
(Cogar et al. 1977, Horkel 1979).
Within the Lake Pasture an unfenced, 522-ha predator-reduction area (PR)
and a 620-ha control area (CO) were chosen due to similarities in: 1) vegetation
consisting of open prairie that was favored nesting habitat (Cogar et al. 1977,
Horkel 1979, Lutz 1979); 2) development for petroleum production and all-
weather roads (Lutz 1979); 3) prairie-chicken populations (34 and 32 males in
1980 and 38 and 31 males in 1981 using leks in PR and CO, respectively), and
4) prairie-chicken nest success in the 2 years immediately prior to this study
(calculated from Lutz 1979: 47-55). Also, during each of the 4 years before this
study, no radio-tagged prairie-chickens were known to travel the 2.3 km between
the most proximate boundaries of these 2 treatment areas (R. S. Lutz, unpubl.
data).
Methods
Striped skunks (N = 74), opossums (N = 83), and raccoons (N = 9) were
removed from PR by trapping and spotlight hunting during 1980 and 1981.
Steel leg-hold traps (Victor 1.5, double-coil spring) and wire-cage live traps
(Havahart) with sardine bait were set during February-June 1980 and 1981 and
1995 Proc. Annu.
Conf.
SEAFWA

Predator Control and Prairie-Chickens 211
checked daily (March-May is the nesting season for Attwater's prairie-chicken).
Leg-hold traps were placed in the end of culverts to avoid interaction with non-
target wildlife and cattle. During 76 nights in 1980, an average of 14 leg-hold
and 9 live traps were employed and during 84 nights in 1981, 12 leg-hold and 9
live-traps were employed. Captured animals were killed immediately. Predators
were further removed by spotlight hunting, shooting mammalian predators with
a .22 caliber rifle along roads and mowed pipeline rights-of-way within the PC
area. Spotlight sampling (Rybarczyk et al. 1980) was conducted to evaluate
predator levels in CO. Biweekly predator indices (predators observed/km) from
March-May were compared between PR and CO using paired Mests for each
year.Prairie-chicken hens were captured using a helinet (Brown 1981) and radio-
tagged with 17-g solar-powered transmitters (Wildl. Materials Inc., Carbondale,
111.) attached with a backpack harness. Radio-marked hens were located 2-7
times per week. Hens were not flushed until they were thought to be incubating
in order to minimize disturbance during laying.
Initiation dates were estimated based upon location data or, if the hatch
date was known, were calculated by back-dating, assuming a laying rate of
1 egg/day and a 23-day incubation period (Horkel 1979:54). Nests in the egg-
laying stages were rarely located and nests destroyed during laying were not
found. We assumed that chronology of nesting should be the same in PR and
CO;
therefore, any difference in initiation date between the 2 areas may reflect
differential nest loss during laying and subsequent renesting. Mean nest initia-
tion was calculated as the sum of the number of days since the first known
initiation of the year to the estimated initiation date for each bird divided by
sample size.
Nest success was calculated by the Mayfield Method (Mayfield 1975). In
using this method, we assumed an average exposure of 35 days (12 days laying
plus 23 days of
incubation).
The Mayfield Method accounts for the higher prob-
ability of hatching in nests discovered nearer to the date of
hatch.
We also calcu-
lated apparent nest success to compare with data collected during earlier
studies.
Hen survival was estimated using the product-limit method (Kaplin and
Meier 1958, PROC LIFETEST, SAS Inst. 1990). We compared homogeneity of
survival curves between areas using Log-rank tests (SAS Inst. 1990).
Results
Predator Removal
In 1980, 106 predators, consisting of 49 striped skunks, 52 opossums, and
5 raccoons, were removed from PR 1 month before and during the nesting pe-
riod (10 Feb-24 Jun) (Table 1). In 1981, 25 striped skunks, 31 opossums, and
4 raccoons were removed during 10 February-27 June. Predator indices from
March-May were lower in PR than in CO both in 1980 (PR x = 0.12 predator/
1995 Proc. Annu.
Conf.
SEAFWA

278 Lawrence and Silvy
Table 1. Numbers of predators removed from the predator-reduction
area by method in the Lake Pasture, O'Connor Brothers' River Ranch,
Refugio County, Texas, 1980 and 1981.
Species
Striped skunk
Opossum
Raccoon
Hunting
38
14
0
1980
Trapping
11
38
5
Hunting
17
12
0
1981
Trapping
8
19
4
Total
74
83
9
km, COx = 0.71; paired Mest, t = 5.99, P < 0.002) and 1981 (PR x = 0.05
predator/km, CO x = 0.42; paired Mest, t = 5.99, P < 0.001).
Attwater's Prairie-Chicken Production
Nineteen nests were found by telemetry and 4 were located by other meth-
ods.
Twelve Attwater's prairie-chicken nests were located during 1980. Two of 3
nests (all first nests) in PR and 2 of 9 in CO (1 known renest) were successful.
All unsuccessful nests were destroyed by predators except 1 in CO, where the
hen was killed 150 m from the nest site, but the eggs were not destroyed.
Seven of 8 nests within PR were successful during 1981. All nests in PR
were thought to be first nests except the
1
unsuccessful nest, which was probably
a renest. Two of 4 nests in CO were successful; both successful nests were first
nests,
while 1 renest was lost to a predator and 1 abandoned due to flooding.
The flooded nest was excluded from this analysis.
Combined 1980 and 1981 nest success was higher (x2 = 5.49, 1 df, P <
0.019) in PR (82%) compared to CO (33%) (Table 2). Mayfield's (1975) nest
success was 75% in PR and 26% in CO. Mean nest initiation dates were signifi-
cantly different (Mann-Whitney U; z = 2.47, 9 and 10 df, P < 0.01) between
PR (5.2 days after the first estimated initiation) and CO (13.4 days after the first
estimated initiation).
Extensive flooding occurred in Lake Pasture following 33.3 cm of rain on
1-3 May 1981 and 25.6 cm on 11-12 June 1981. Increased adult mortality oc-
curred during the week following the 1-3 May storm. During this period, 6 of
7 radio-tagged hens with broods died while 1 of 5 hens nesting or without
broods died. No radio-monitored broods were known to survive the 1-3 May
storm nor were any broods observed later in the year. Only 1 of 3 incubating
hens abandoned her nest as result of the flooding; the 2 other hens continued
to incubate after their nests had been temporarily inundated.
Radio-marked hen survival was low during the breeding season (Fig. 1).
We combined years for analysis since there was no difference in survival between
years (Log-rank x2 = 0.13, 1 df, P < 0.716). The survival curves were different
between areas (Log-rank x2 = 5.96, 1 df, P < 0.015); however, calculated sur-
vival was low in both PR (survival = 0%) and CO (survival = 9%). Sixty-four
1995 Proc. Annu.
Conf.
SEAFWA

Predator Control and Prairie-Chickens 279
CO
0 10 20 30 40 SO 60 70
Days
Fig. 1. Product-limit survival estimates
for radio-marked Attwater's prairie-chickens
in Refugio County, Texas, during 10 March-
30 June 1980 and
1981,
CO N = 18 and
PR
TV
= 25.
percent (N = 25) of mortalities, where the cause was known, were attributed to
mammalian predators, 16% were due to avian predators, and 20% were
weather-related.
Discussion
Predator indices and number of predators removed from PR suggest that
nest predator populations in PR were lower than CO. However, Lutz (1979)
noted that predator activity, as indexed by coyote, skunk, and raccoon scat
counts along roads, ranked lower in PR than CO in 1978-79. The original study
design called for switching the predator reduction and control areas during the
second year of
the
study; however, the small sample of nests in PR during 1980
and concern about residual effect of predator reduction resulted in our continu-
ing with the same areas during the second year. This weakened our inferences
about the effect of the reduction program on predator numbers. There were
indications that predator numbers in PR were lower in 1981 than 1980, sug-
gesting a residual effect (Lawrence 1982: 49); however, Lutz (1979: 64) noted a
significant difference in predator indices in PR between years when predator
reduction was not in effect. Duebbert and Kantrud (1974) and Greenwood
(1986) noted that the residual effects of predator control were of short duration.
Beasom (1974) noted coyotes and bobcats (Lynx
rufus)
increased rapidly to for-
mer levels after completion of short-term predator-control program, though in-
dices for smaller predators were erratic throughout his study.
We believe that by removing skunks, opossums, and raccoons, we were
focusing on the major nest predators. Although coyotes (Canis
latrans)
were
known to prey on Attwater's prairie-chicken nests, no attempts were made to
1995 Proc. Annu.
Conf.
SEAFWA

280 Lawrence and Silvy
control coyotes because
of
the small study areas.
A
few eggs are taken by snakes
(Lutz 1979),
but no
other nest predator was observed taking Attwater's prairie
chicken eggs during the
4
years prior
to our
study.
Removal
of a
large number
of
potential nest predators from
PR
apparently
resulted
in
increased nest success (82%
on PR
vs. 33%
on
CO). Because treat-
ments were
not
replicated, treatment and site effects may
be
confounded. How-
ever, data collected
on
these study areas during 1978-79 (calculated from Lutz
1979:47-56), indicated 32% success
for
nests
(N = 22) on the
area that later
became
our
control site
and
35%
(N = 17)
success
for
nests
on the
area that
became our predator reduction site (Table 2). Lehmann (1941:37) observed 32%
(N = 19) nest success
for
Attwater's prairie-chickens
in
Colorado County, Texas,
during 1937-38
and
Horkel (1979:55), working
on our
study area during 1976—
77,
observed nest success
of
42%
(N =
19). The later mean nest initiation date
for hens
in CO
probably resulted from greater nest destruction during
egg-
laying. This hypothesis
is
compatible with
the
higher nest predation observed
in CO.
Breeding season
hen
survival
was
lower than
the
36% reported
by
Lutz
et al. (1994), data that included
the
sample
of
hens from CO during this study.
The shape
of
the survival curves was different between the
2
areas; however,
by
30 June survival was low
in
both areas. Survival estimates may have been biased
low (Burger
et
al. 1991); however, transmitters
and
attachment techniques were
the same
as
used during other studies
on
Attwater's prairie-chickens where sur-
vival
was
greater (Lutz
et al.
1994). Potential predators (coyotes, white-tailed
hawks [Buteo albicaudatus],
and
great horned owls [Bubo virginianus])
of
adult
Attwater's prairie-chickens were observed on both areas, and
a
great horned owl
nested
in PR
during 1981. Jackrabbit (Lepus californicus) and eastern cottontail
(Sylvilagus floridanus) indices were
3.5
times greater
in CO
than
in PR
during
1981 (Lawrence 1982). Coyotes
and
avian predators
may
have responded
to
lower densities
of
skunks
and
opossum
in PR by
increasing predation on jack-
rabbits, cottontails,
and
adult prairie-chickens. While most
hen
mortality
was
attributed
to
mammalian predation,
the
figure
for
mammalian predation
may
be biased upwards since some carcasses may have been scavenged.
Table 2. Nest success of Attwater's prairie-chicken
in
predator-reduction area (PR) and control areas (CO)
in
the Lake Pasture, O'Connor Brothers' River Ranch,
Refugio County, Texas, 1978-1979 (Lutz 1979,
no
predator reduction) and 1980-1981 (our study).
Areas
PR
CO
N nests
17
22
1978-1979
%
successful
35
32
N nests
11
12
1980-1981
%
successful
82
33
1995 Proc. Annu.
Conf.
SEAFWA

Predator Control
and
Prairie-Chickens
281
Predation
was not the
only variable that influenced prairie-chicken
num-
bers during
our
study. Flooding dramatically affected productivity during
1981.
Following
the
first major storm,
we
observed increased mortality
of
radio-
tagged hens.
A
second major storm apparently eliminated
any
production that
may have resulted from renesting,
as no
broods were observed after
1 May.
Lehmann (1941:32-35)
and
Horkel (1979:88) also documented loss
of
nest
and
broods
of
Attwater's prairie-chicken
to
heavy rains.
Management
of an
endangered species, such
as
Attwater's prairie-chicken,
might require manipulation
of
predator population levels
to
favor survival,
es-
pecially considering
the
island nature
of
many habitats that
may
concentrate
both prairie-chickens
and
predators.
We
were unable
to
draw strong inferences
about
the
effect
of
predator reduction
on
prairie-chicken nest success; small
sample size
of
nests
and
unreplicated treatments make
our
findings equivocal.
However, this study will
not be
replicated
due to
current
low
numbers of Attwat-
er's prairie-chickens
in the
wild. These results
may
help managers assess
the
risk
of alternative management techniques
to
increase Attwater's productivity
in the
wild. There
is
some indication that adult survival
was
lower
in the
predator
reduction area,
and it may be
necessary
to
control
a
diverse group
of
species
that prey
on
both prairie-chicken adults
and
nests
for a
control program
to
be effective.
A predator-reduction program used
in
conjunction with other management
tools such
as
electric fences around nests
to
increase nest success (Lokemoen
et
al.
1982)
might prove beneficial
for
Attwater's prairie-chicken.
If
techniques,
habitats,
and
captive-reared prairie-chickens become available
for
reintroduc-
tion into
the
wild, predator reduction
may be
used
at the
release site
to
increase
the probability
of a
successful transplant.
Literature Cited
Anderson,
W. 1957. A
waterfowl nesting study
in the
Sacramento Valley, California,
1955.
Calif.
Fish
and
Game 43:71-90.
Balser,
D. S., H. H.
Dill,
and H. K.
Nelson. 1968. Effect
of
predator reduction
on
water-
fowl nesting success.
J.
Wildl. Manage. 32:669-682.
Beasom,
S. L. 1974.
Intensive short term predator removal
as a
game management tool.
Trans.
North
Am.
Wildl.
Nat.
Resour.
Conf.
39:230-240.
Braun,
C. E., K. W.
Harmon,
J. A.
Jackson,
and C. D.
Littlefield.
1978.
Management
of
national wildlife refuges
in the
United States:
its
impact
on
birds. Wilson Bull.
90:309-321.
Brown,
D. L.
1981.
The
helinet:
a
device
for
capturing prairie chickens
and
ring-necked
pheasants. Proc. Annu.
Conf.
Southeast. Assoc. Fish
and
Wildl. Agencies 35:92-96.
Burger,
L. W., M. R.
Ryan,
D. R
Jones,
and A. R
Wywialowski. 1991. Radio transmitters
bias estimation
of
movements
and
survival.
J.
Wildl. Manage. 55:693-697.
Chesness,
R. A., M. M.
Nelson,
and W. H.
Longley. 1968.
The
effect
of
predator reduc-
tion
on
pheasant reproductive success.
J.
Wildl. Manage. 32:683-697.
Cogar,
V. F, J. D.
Horkel,
and N. J.
Silvy. 1977. Vegetative type preference
of
Attwater's
1995 Proc. Annu.
Conf.
SEAFWA

282 Lawrence
and
Silvy
prairie chicken
in
coastal prairie. Proc. Annu.
Conf.
Southeast. Assoc. Fish
and
Wildl. Agencies 31:41-50.
Duebbert,
H. F. and H. A.
Kantrud. 1974. Upland duck nesting related
to
land
use and
predator reduction.
J.
Wildl. Manage. 38:257-265.
Greenwood,
R. J.
1986. Influence
of
striped skunk removal
on
upland duck nest success
in North Dakota. Wildl. Soc. Bull.
14:6-11.
Horkel,
J. D.
1979. Cover
and
space requirements
of
Attwater's prairie chicken
(Tympa-
nuchus cupido attwateri)
in
Refugio County, Texas. Ph.D. Diss., Texas A&M Univ.,
College Station. 96pp.
,
R. S.
Lutz,
and
N.
J.
Silvy. 1978.
The
influence
of
environmental parameters
on
nesting success
of
upland game birds. Proc. Annu.
Conf.
Southeast. Assoc. Fish
and Wildl. Agencies. 32:234-241.
Kaplan,
E. L. and
P.
Meier. 1958. Nonparametric estimation from incomplete observa-
tions.
J. Am.
Stat. Assoc. 53:457-481.
Lawrence, J. S. 1982. Effect
of
predator reduction
on the
reproductive success
of
Attwat-
er's prairie chicken. M.S. Thesis, Texas A&M Univ., College Station. 105pp.
and N.
J.
Silvy. 1980. Status
of
the Attwater's prairie chicken—an update. Pages
29-33
in
P.
A.
Vohs
and F. L. Knopf, eds.
Proc. Prairie Grouse Symp. Stillwater,
Okla.
Lehmann,
V.
W.
1941.
Attwater's prairie chicken,
its
life history
and
management. North
Am. Fauna 57. 63pp.
. 1968.
The
Attwater's prairie chicken, current status
and
restoration opportuni-
ties.
Trans. North Am. Wildl.
Nat.
Resour.
Conf.
33:398-407.
Livezey,
B.
C.
1981.
Duck nesting
in
retired croplands
at
Horicon National Wildlife
Ref-
uge,
Wisconsin.
J.
Wildl. Manage. 45:27-37.
Lokemoen,
J. T, H. A.
Doty,
D. E.
Sharp,
and J. E.
Neaville.
1982.
Electric fences
to
reduce mammalian predation
on
waterfowl nests. Wildl. Soc. Bull. 10:318-323.
Lutz,
R.
S. 1979.
The
response
of
Attwater's prairie-chicken
to
petroleum development.
M.S.
Thesis, Texas A&M Univ., College Station. 82pp.
,
J. S.
Lawrence,
and
N.
J.
Silvy.
1994.
Nesting ecology
of
Attwater's prairie-
chicken.
J.
Wildl. Manage. 58:230-233.
Mayfield,
H.
1975. Suggestions
for
calculating nest success. Wilson Bull. 87:456-466.
Rybarczyk,
W.
B.,
R. D.
Andrews,
E. E.
Klass,
and J. M.
Kienzler. 1980. Raccoon spot-
light survey technique:
a
potential population trend indicator. Pages 1413-1430
in
J. A.
Chapman
and D.
Pursley, eds. Worldwide Furbearer
Conf.
Proc.
R. R. Don-
nelley
and
Sons Co., Falls Church,
Va.
Sargeant,
A.
B.
and
P.
M.
Arnold. 1984. Predator management
for
ducks
on
waterfowl
production areas
in the
northern plains. Proc. Vertebrate Pest
Conf.
11:161-167.
SAS Institute
Inc. 1990.
SAS/STAT User's Guide, Version
6,
Fourth
Ed., Vol. 2, SAS
Inst.
Inc.
Cary, N.C. 846pp.
Stoddard,
H. L. and E.
V.
Komerek.
1941.
Predator control
in
southeast quail manage-
ment. Trans. North Am. Wildl.
Nat.
Resour.
Conf.
6:288-293.
Trautman,
C. G, L. G.
Fredrickson,
and A.
V.
Carter.
1974.
Relationship
of red
foxes
and other predators
to
populations
of
ring-necked pheasants
and
other prey, South
Dakota. Trans. North Am. Wildl.
Nat.
Resour.
Conf.
39:241-255.
1995 Proc. Annu.
Conf.
SEAFWA