ArticlePDF Available

Effectiveness of barriers, pyrotechnics, flashing lights, and Scarey Man for deterring heron predation on fish

Man Fall
Guy. Photo
by William Andelt. Great blue heron. Photo
by William Andelt.
F. Andelt, Timothy P. Woolley, and Stuart N. Hopper
Pyrotechnics are effective, but flashing lights and Scarey Man? are
ineffective in deterring heron predation on trout
Authors' address
this research:
of Fishery
Colorado State
CO 80523,
USA. Current address for
P. Woolley:
Game and Fish
P.O. Box 111, Savery,
82332, USA.
words: Ardea
frightening, great
blue herons,
Society Bulletin 1997, 25(3):686-694
Peer edited
heron predation
on fish * Andelt
et al. 687
Fish-eating birds are significant predators at fish-
culture facilities (Scanlon et al. 1978, Parkhurst et al.
1987, Stickley and Andrews 1989, Stickley et al.
1995a). Black-crowned night-herons (Nycticorax
nycticorax; BCNH) and great blue herons (Ardea
berodias; GBH) have been estimated to capture an
average of 79,000 rainbow trout (Oncorhynchus
mykiss) annually from 1985 to 1989 at the Watson
Lake Fish Rearing Unit near Bellvue, Colorado (D. C.
Smeltzer, Colo. Div. Wildl., Bellvue, pers. commun.).
Several techniques have been used or recom-
mended to reduce heron predation on fish in hatch-
eries (Salmon and Conte 1981, Draulans 1987,
Parkhurst et al. 1987). Parkhurst et al. (1987) re-
ported limited success with hand-fired
but Stickley and Andrews (1989) observed that py-
rotechnics were the most effective technique used
by Mississippi farmers
for frightening fish-eating
from catfish ponds. Naggiar (1974) reported that
flashing lights with reflectors were effective during a
short period, for deterring BCNH and GBH foraging
at night. However, pyrotechnics and oscillating
lights have not been thoroughly evaluated.
(1939) and Parkhurst et al. (1987) reported
limited to intermediate success with scarecrows in
frightening herons from hatcheries. A relatively new
technique, the Scarey Man' Fall Guy (R. Royal, P.O.
Box 108, Midnight, MS 39115), a 1.46-m-tall,
mannikin that intermittently
inflates with air, bobs up
and down, illuminates with an internal light, and
emits a high-pitched wail (Stickley et al. 1995b), has
been developed but has not been evaluated for fright-
ening BCNH and GBH. Stickley
et al. (1995b) reported
that Scarey Man reduced numbers of double-crested
cormorants (Phalacrocorax auritus) at catfish ponds
in Mississippi,
but the birds became habituated to the
devices at 3 of 4 sites. We evaluated the effectiveness
of nylon-mesh barriers,
pyrotechnics, bright rotating
(flashing) lights, and Scarey
Man for reducing number
of herons and predation on fish at a rearing unit near
Bellvue, Colorado.
Study area
The study was conducted at the Colorado Division
of Wildlife's Watson Lake Fish Rearing Unit (Fig. 1)
near Bellvue, Colorado. The raceways were formed
by concrete walls averaging 1.30 m high and extend-
ing about 11-18 cm above the ground. The water in
the raceways averaged 70 cm (range = 49-89 cm)
deep and 60 cm (range = 35-83 cm) below the top of
the raceway walls. Eight steel-mesh crosswalks and
140 fish feeders were located on the raceways (Fig.
1). The number of trout (10-23 cm long) maintained
I - I
T - .- T -
Fl -H
I rI_
- !
; 1 -I- 1 -- I
[~~~ II. I H I I
f~~~~ I 7
0 Raceway
Fig. 1. Location of raceways, raceway walls, fish feeders, nylon-
mesh barriers behind feeders, crosswalks, and Scarey Man (black
rectangles), at the Watson Lake Fish Rearing
Unit in Colorado dur-
ing 1990-1992.
in the raceways during periods when herons were
present was similar during 1990-1992 and ranged
from about 320,000 in April to 120,000 in July.
Methods and materials
We observed herons (Table 1) with lOx binoculars
or a Javelin night-viewing device (avelin Electronics,
Los Angeles, Calif.) equipped with a 300-mm tele-
photo lens from a vehicle on a public road 2 m higher
than and 35-58 m north of the raceways. Hatchery
personnel attached nylon netting (2-mm diameter
and 7- x 7-cm mesh, 2.3 m long, 0.5 m high, and
slanted toward the water at 30-40? from vertical) be-
hind the fish feeders in 1991 and 1992, but not in
1990, to deter herons from capturing fish (Fig. 1).
We hid in a blind 10 m south of the raceways and
fired pyrotechnics at herons immediately after they
landed. We used a single-shot pistol (Record PTB
113, imported through Sutton Ag Enterprises,
Calif.) to fire 17-mm exploding projectiles (Pest Con-
trol Devices, Stoneco, Inc., Trinidad, Colo.) from
2000 to 2200 hours and 15-mm whistling projectiles
(Bird Whistlers', Zink-Feuerwerk, GmbH, Clee-
bronn, Germany)
from 2200 to 0700 hours. We used
the exploding projectiles before 2200 hours because
we surmised that they would more effectively frighten
herons, whereas the whistling projectiles were used
after 2200 hours to limit disturbances to neighbors.
The projectiles flew about 40-50 m.
I - I - * it I I - -
-- i
688 Wildlife
Table 1. Chronology
of research conducted on black-crowned
and great
blue herons
at the Watson
Lake Fish
Unit in Colorado
1990, 1991, and 1992.
Nighta Number
of nights
activity Beginning Ending Observing Frightening
Pretreatment counts May
19-20 May
21-22 3
fired for
14 nights May
22-23 Jun
4-5 14
Posttreatment and
counts Jun 5-6 Jun
26-27 8
night Jun
26-27 Jul
29 14 7
Posttreatment counts Aug
2 Sep
25 6
Pretreatment counts Jun
15b Jun
19-20b 6
Rotating lights Jun
21-22C Jul
14-15c 8 24
Posttreatment and
counts Jul 16 Jul
22 4
for 7 nights Jul
23-24 Jul
29-30 7
counts Jul 31 Aug
21 7
1 May
26-2 7 May
31-1 Jun 6
1 Jun 1-2 Jun
18-19 8 18
2 Jun
19-20 Jun
28-29 5
ScareyMan2 Jun 29-30 Jul
12-13 6 14
Posttreatment 2 Jul 13-14 Jul
22-23 5
Two dates
the month
indicate observation
the night;
1 date indicates observation
the early
the night.
from 0200 to 0700 on 15, 16, 17, and 18 June,
2200 to 0700 on 18-19 June,
and 2100 to 0700 on 19-20 June.
2100 to 0700 on 21-22 and 22-23 June,
0200 to 0700 on 26 and 29 June,
2, 6, 10, and
14 July.
Counts occurred
2100 to 0200 during
all 5 periods.
We placed 4 2-beam rotating beacons with clear
acrylic domes and 75,000 candle-power spotlamps
(Tripp Lite
Model Mark-12,
Tri Lite, Inc., Chicago, Ill.)
at the 4 corners of the western half of the rearing
unit. Plywood was placed immediately east of the 2
eastern lights to minimize flashing into the eastern
half of the rearing unit. One 2-beam rotating beacon
with a clear acrylic dome (Tripp Lite Model RF-6,
Lite, Inc., Chicago, Ill.) and 70,000 candle-power
halogen spotlamps (No. H7616, Wagner Div., Cooper
Industries, Inc., Parsippany, NJ.) was positioned in
the center of the western half of the rearing unit.
Each beacon rotated 45 times and produced 90
flashes per minute.
We used the night-viewing device to count herons
on the eastern half of the rearing
unit during pretreat-
ment, treatment, and posttreatment periods, and on
the western half during pretreatment and posttreat-
ment of the flashing-light experiment. We used
binoculars to count herons on the western half of the
rearing unit while lights were flashing (treatment pe-
riod) because bright lights could have damaged the
night-viewing device. To determine if we biased our
counts by using different viewing devices, we painted
plastic, beverage bottles to simulate juvenile and
adult BCNH and placed them in the rearing unit; our
observer counted similar proportions of bottles on
lighted (24/26) and unlighted (20/26) sides.
We evaluated the effectiveness of 2 Scarey Man
mannikins by conducting hourly counts of herons
during 5 consecutive periods (Table 1). The man-
nikins were placed at the northwestern site on the
east half and at the southwestern site on the western
half of the rearing unit (Fig. 1) during the first 2
nights and then rotated counter-clockwise on each
half of the rearing unit every other night to minimize
habituation. Each Scarey Man activated for 35-40
seconds every 9-10 minutes from about 1700 or
2045 through 0800 hours. To evaluate habituation to
Scarey Man, we determined the proportion of BCNH
and GBH
that remained on the half of the rearing
where a Scarey Man activated; we divided the num-
ber of BCNH and GBH
observed <2 minutes after the
Scarey Man activated by the number of BCNH ob-
served <2 minutes before it activated.
Data analyses
The effect of nylon-mesh barriers on altering the
distribution of BCNH and GBH behind feeders was
determined by comparing the nightly proportions of
BCNH and GBH behind feeders without barriers in
1990 to the nightly proportions of BCNH and GBH
heron predation
on fish * Andelt
et al. 689
behind feeders with barriers
in 1991 and 1992. The
comparisons were made with ANOVA (GLM proce-
dure) after weighting the nightly proportions by the
total number of herons observed in the rearing unit.
We used linear regression (GLM procedure, SAS
Inst., Inc. 1988:549-640) to examine the effect of
time (nights) on the number of pyrotechnics fired,
number of BCNH and GBH frightened during py-
rotechnic experiments, number of BCNH and GBH
that returned after the 14- and 7-night pyrotechnic
experiments, number of herons present during all 5
periods of the Scarey Man experiment, and propor-
tion of herons present <2 minutes after each Scarey
Man activated. The average numbers of herons pres-
ent each night before and after the 14- and 7-night-py-
rotechnic experiments were compared with t-tests
(TTEST procedure, SAS Inst., Inc. 1988:941-947). If
the variances were unequal, an approximate t was
calculated. Satterwaite's (1946) approximation was
used to compute the degrees of freedom and the
probability level for the approximate t. When py-
rotechnics were fired every fifth night, the numbers
of herons observed the nights before and after fright-
ening were compared with a paired t-test (MEANS
procedure, SAS Inst., Inc. 1988:946-947).
The effect of flashing lights on herons was deter-
mined by subtracting the number of herons observed
per hour each night on the eastern half of the rearing
unit from the number observed on the western half
and then comparing the differences among pretreat-
ment, treatment, and posttreatment with a paired
analysis of variance (ANOVA; GLM procedure, SAS
Inst., Inc. 1988:549-640). The average number of
herons observed each night at the entire rearing unit
during pretreatment, treatment, and posttreatment of
the flashing-light and Scarey Man experiments were
compared with ANOVA. We conducted this research
following animal welfare protocol number 90-071,
which was approved by the Colorado State Univer-
sity Animal Care and Use Committee.
Diurnal and seasonal activity patterns
BCNH and GBH arrived at the rearing unit between
2000 and 2100 hours, and most birds of both species
departed by 0700 hours. BCNH and GBH started ar-
riving in April (D. C. Smeltzer, Colo. Div. Wildl., Bell-
vue, pers. commun.) and only a few of either species
remained by mid-to-late September.
The nightly proportions of BCNH counted behind
feeders, compared to all areas, were similar (Contrast
F = 0.09; 1, 80 df; P = 0.765) before (2%
of 2,215 ob-
servations during 1990) and after (2% of 11,579 ob-
servations during 1991 and 1992) barriers
were pres-
ent. The proportions of GBH
counted behind feeders
varied (F = 76.70; 2,79 df; P < 0.0001) among years
and were higher during 1990 (32% of 584 observa-
tions) without barriers compared to 1991 (10% of
1,356 observations) and 1992 (4% of 1,521 observa-
tions) with barriers. The barriers were associated
with an increase in hourly counts of GBH
at the rear-
ing unit.
Frightening with pyrotechnics for 14
The average number of pyrotechnics fired (x =
4.0, SE = 0.5) and BCNH (x = 3.1, SE = 0.7) and GBH
(x = 2.0, SE = 0.3) frightened per hour decreased
from nights 1 through 14 of the experiment (Fig.
2a). The average number of BCNH (x = 55.9, SE =
12 -- -.
03 8
2c 6
I 4
U) cO
0 4
cc 2
X1 l I I
I i
i , I I I I I . I
1 2 3 4 5 6 7 8 9 10 11 12 13 14
0 -
rO H
t0 - BCNH
= 0.63 + 0.45
x night,
r2 = 0.79,
P = 0.003
= 0.47 + 0.15
x night,
r2= 0.78,
P= 0.004
2L\. _
3 2 1 1 2 4 7 10 14 18 22
2. (a)
number of pyrotechnics
and great blue herons (GBH)
frightened per hour (2100-0700) during
14 consecutive nights
(22-23 May to 4-5 Jun 1990) at the Watson
Lake Fish
Unit in Colorado. (b) Average
numbers of BCNH
and GBH
per hour
(2100-0700) before
(19-20 to 21-22 May
and after
(5-6 to 26-27 Jun
firing pyrotechnics
for 14 nights.
- A
= 6.98
- 0.39
x night,
r2 = 0.68,
P = 0.0003
= 7.07
- 0.53 x night,
r2 0.66,
P = 0.0004
mma,A GBH = 3.26 - 0.17 x night, r2 = 0.44, P= 0.010
.^..v . '
i J ! , , , , , , , , , ,
690 Wildlife
Society Bulletin
1997, 25(3):686-694
9.4) and GBH (x = 7.0, SE = 1.0) observed per hour
each night before firing pyrotechnics for 14 consec-
utive nights was higher (BCNH:
t = -5.4, 2.1 df, P =
0.031; GBH: t = -5.3, 9 df, P = 0.0005) than the av-
erage number (BCNH: x = 5.0, SE = 1.4; GBH:
x =
1.9, SE = 0.5) observed after frightening (Fig. 2b).
After firing pyrotechnics, the average number of
BCNH and GBH increased slowly from nights 1
through 22 (Fig. 2b).
Frightening with pyrotechnics for 7
The average number of pyrotechnics fired (x =
1.2, SE
= 0.2) and BCNH
(x = 1.5, SE
= 0.3) and GBH
(x = 0.3, SE = 0.1) frightened per hour did not
change from nights 1 through 7 (Fig. 3a). The aver-
age number of BCNH
(x = 10.4, SE
= 1.7) and GBH
= 4.7, SE = 1.4) observed per night after firing py-
-% Pyrotechnics = 1.90 - 0.17x night, r2 = 0.42, P= 0.113
\ BCNH = 2.47 - 0.24 x night, r2 = 0.33, P = 0.181
% GBH = 0.53 - 0.05 x night, r = 0.23, P= 0.276
40 - l -----.---
r 1 BCNH
= 6.00 + 0.51 x night,
r2 = 0.71,P = 0.017
GBH = 0.60 + 0.47x night, r2= 0.89, P= 0.001
g g ;;40k1k40 |
3. (a)
number of pyrotechnics
and great blue herons (GBH)
(2100-0700, except
2100-0530 on 27-28 Jul
1991) during
7 consecutive
(23-24 to 29-30 Jul 1991) of
herons from
the Watson Lake
in Col-
orado. (b) Average
numbers of BCNH and GBH
hour (0200-0700) before (16-22 Jul 1991) and after
(31 Jul-21
Aug 1991) firing
for 7 nights.
8 7 5 2 1 2
NIGHT 4 7 10 15 22
rotechnics was lower (BCNH: t = -6.46, 9 df, P = <
0.0001; GBH t = -1.78, 9 df, P = 0.109) than the av-
erage number (BCNH x = 29.0, SE = 2.3; GBH x =
8.7, SE = 1.5) observed before frightening (Fig. 3b).
After firing pyrotechnics, the average number of
BCNH and GBH increased from nights 1 through 22
(Fig. 3b). A lower proportion (posttreatment di-
vided by pretreatment) of BCNH and GBH were ob-
served after 14 nights of frightening during 1990
than after 7 nights of frightening during 1991 (Figs.
2b, 3b).
Frightening with pyrotechnics every
fifth night
The average number of pyrotechnics fired (x = 3.9,
= 0.3) and BCNH
(x = 3.5, SE
= 0.3) and GBH
(x =
1.1, SE
= 0.2) frightened per hour did not change dur-
ing 7 consecutive nights (separated by 4 nights with-
out frightening; Fig. 4a). The average number of
( = 16.2, SE = 2.2) and GBH
(x = 5.9, SE = 1.1)
observed per hour each night before frightening was
greater (BCNH:
paired t = 5.6, 6 df, P = 0.001; GBH:
paired t = 3.4, 6 df, P = 0.014) than the average num-
ber (BCNH:
x = 6.0, SE = 1.0; GBH:
x = 2.2, SE
= 0.7)
observed per hour each night after frightening (Fig.
4b). The average number of BCNH and GBH in-
creased during the treatment (nights before frighten-
ing: BCNH:
x = 16.9; GBH:
x = 5.8; Fig. 4b) compared
to an average of 7.9 BCNH and 4.6 GBH on nights 5,
9, and 13 before treatment. This was likely related to
a continuing increase in herons following the 14
nights of frightening with pyrotechnics in the previ-
ous experiment. The average number of BCNH
(counted at 0400 and 0500 hrs) continued to increase
on nights 5 (x = 31.0) and 11 (x = 50.5) after treat-
ment and then decreased on nights 23 (x = 32.0), 35
(x = 16.0), 47 (x = 5.5), and 59 (x = 0.5) after treat-
ment. The average number of GBH
(counted at 0400
and 0500 hrs) also increased or remained similar to
treatment numbers on nights 5 (x- = 5.5), 11 (x =
10.5), 23 (x = 3.5), 35 (x = 7.5), and 47 (x = 5.0), but
decreased on night 59 (x = 1.5) after treatment. The
decreases in numbers of BCNH and GBH during the
latter posttreatment period likely was related to mi-
gration from the study area.
Bright flashing lights
The number of BCNH on the western half of the
rearing unit was higher than on the eastern half dur-
ing pretreatment (western x = 16.0, SE
= 1.5; eastern
x = 12.3, SE = 0.9) and posttreatment periods (west-
ern x = 16.5, SE = 1.3; eastern x = 12.4, SE = 1.5),
whereas the number of BCNH was lower on the
lighted western half of the rearing unit (x = 12.9, SE
, 30
z In g
*g I P
i I W ii r ' ** * -
I -
on fish * Andelt
et al. 691
23 .v. BCNH = x night,
r2 =
1 2 3 4
oc a:
I 25
I 20
w 10
m 5
F - V
,ir if ~
F t 7
1 2 3 4
4. (a) Average
number of pyrotechnic
and gre
per hour (2100-0700) during;
27-28 Jul
each separated
4 nights
fka XA/ti-crn I l-^o
F:ick Pn3rin: I Init in
:- nlnr
t, r2 0.27, P= 0.228 4.
P - 0.355
P= 0.114
treatment (x = 29.4, SE = 3.4), and posttreatment (x
= 28.9, SE = 2.3) periods (F = 0.04; 2,15 df; P =
0.958). The average number of GBH on the western
and eastern halves of the rearing unit combined was
greater during pretreatment (x = 11.8, SE = 0.8) than
during treatment (x = 8.5; SE = 0.8) and posttreat-
ment (x = 8.7, SE = 1.5) periods (F = 3.87; 2,15 df; P
= 0.044).
Frightening with Scarey Man
The average number of BCNH at the rearing unit
5 6 7 varied (F = 16.00; 4,25 df; P < 0.0001) among the 5
pretreatment, treatment, and posttreatment periods
(Fig. 6), and was lower during the first treatment than
B during the first pretreatment or first posttreatment
periods (Table 2). The number of BCNH also was
lower during the second treatment and second post-
treatment periods than during the second pretreat-
I ment period. The average number of GBH varied (F
= 4.55; 4,25 df; P = 0.007) among the 5 periods and
generally was lowest during treatment periods (Fig.
7 ( j 6, Table 2). The average numbers of BCNH and GBH
/ j( I increased with time (nights) during the first treat-
:t '? g ment period, but did not change with time during the
other periods (Table 2).
5 6 7 The proportion of BCNH that remained (<2 min af-
ter frightening) on the half of the rearing unit where
a Scarey Man activated increased with time (propor-
at blue herons (GBH) tion BCNH = 0.66 + 0.03 x night, n = 75, r2 = 0.26, P
7 nights
(27-28 Jun
to < 0.0001) from day 1 (n = 13, x no. before = 6.9, x
at no. after = 4.1) through day 14 (n = 12, x no. before
-Ar . !. A/, rrn\ n... ....
1 VVdLUI I LdlKe r
Is INerdlIg I J IIIL II I ,UIUIdUU. tU) Averadge inumI-
bers of BCNH
and GBH observed per hour (0100-0200 and
0400-0600) the night
and the night
firing pyrotech-
= 1.7) than on the unlighted eastern half (. = 16.5, SE
= 1.7) during treatment (paired ANOVA, F = 19.52;
2,15 df; P < 0.0001; Fig. 5). Similarly, the number of
GBH on the western half of the rearing unit was
higher than on the eastern half during pretreatment
(western x = 7.3, SE = 0.4; eastern x = 4.5, SE = 0.5)
and posttreatment periods (western x = 5.2, SE = 0.8;
eastern x = 3.5, SE = 0.9), whereas the number of
GBH was lower on the lighted, western half of the
rearing unit (x = 2.3, SE = 0.3) than the unlighted
eastern half (x = 6.2, SE = 0.7) during treatment
(paired ANOVA, F = 32.34; 2,15 df; P < 0.0001; Fig.
5). These changes in numbers of herons indicate that
some of the 21% of BCNH and 63% of GBH were re-
distributed from the western half to the eastern half
of the rearing unit while the lights were flashing.
However, the average number of BCNH on the com-
bined western and eastern halves of the rearing unit
did not vary among pretreatment (x = 28.2, SE = 2.2),
a 20
0 15
W 10
0 7 6 5 4 3 2 1 2 5 8 11 151923 1 2 4 7
5. Average
numbers of black-crowned
and great
blue herons
per hour
(0200-0700) on
the western
and eastern
of the Watson Lake
Unit in Colorado before
(15-20 Jun
1991, pretreatment,
as no. of nights
flashing lights
on the western
while the lights
were flashing
(22 Jun-14
1991, treatment
counts, but lights flashed 21-22 Jun through
14-15 Jul 1991), and after
the lights
were removed
(16-22 Jul
1991, posttreatment,
denoted as number of nights
_ _ I? __ _L_ II I
11 L [
f .
I I i -- - I I -R
f~ LI
i I I
692 Wildlife
Society Bulletin
1997, 25(3):686-694
? 80
co 60
o 40
m 20
~_ C~~~
_ ~i
- rr
' I
i v I i
6 4 2 2 7 14 18 1 4 10
NIGHT 2 7 14 1 4 10
6. Average
of black-crowned
and great
blue herons
per hour
(2100-0200 hr)
(26-27 May
to 31 May-1
1992, pretreatment,
as no. of nights before treatment
1), during
(1-2 to 18-19 Jun
1992, treatment
(19-20 to 28-29 Jun
1992, posttreatment
2, denoted as no. of nights
1), dur-
ing (29-30 Jun
to 12-13 Jul
1992, treatment
2), and after
to 22-23 Jul
1992, posttreatment
2, denoted
as no. of nights
2) 2 Scarey
Man mannikins were used to frighten
from the Watson Lake
in Colorado.
= 16.8, x no. after = 17.7) of the first treatment pe-
riod. The proportion of GBH that remained on the
half of the rearing unit where a Scarey Man activated
did not increase with time (proportion GBH = 0.96 +
0.002 x night, n = 53, r2 = 0.00, P = 0.866) from day
1 (n = 6, x no. before = 2.3, x no. after = 1.5) through
day 14 (n = 12, x no. before = 3.2, x no. after = 3.0)
of the first treatment period.
Estimates of trout captured by herons
We estimate that BCNH and GBH captured a mini-
mum average of 31,000 and 25,000 trout, respec-
tively, annually during 1990, 1991, and 1992 (Andelt,
unpubl. data). These captures occurred while we
frightened herons with pyrotechnics, flashing lights,
and Scarey Man during some periods that herons
were present. Hatchery personnel (D. C. Smeltzer,
Colo. Div. Wildl., Bellvue, pers. commun.) estimated
that BCNH and GBH combined captured an average
of 40,000 trout annually during 1990, 1991, and
1992. These losses of fish to herons were lower than
the estimated average annual losses (79,000) of fish
that hatchery personnel attributed to herons during
the previous 5 years when frightening devices were
not used.
BCNH and GBH were the only avian species that
we observed capturing trout at the Watson Lake Fish
Rearing Unit. In contrast, Parkhurst et al. (1992) re-
ported 7 major avian predators of fish at hatcheries in
Pennsylvania and reported that GBH were not ob-
served foraging in concrete raceways.
The presence of barriers behind fish feeders was
associated with a significant reduction in the number
of GBH that were behind feeders and an increase in
the number of GBH at the rearing unit. Great blue
heron fish-capture rates were highest from behind
fish feeders (Andelt, unpubl. data); this suggests that
GBHs were forced to stay longer at the rearing unit to
capture fish after barriers were installed.
We were more successful in decreasing numbers
of BCNH and GBH by firing pyrotechnics for 14 com-
pared to 7 consecutive nights, and we speculate that
more herons adopted alternate food sources during
the longer period of frightening. Although frighten-
ing herons every fifth night decreased the number of
herons the following night relative to the night be-
fore frightening, the lack of reduction in herons with
each subsequent fifth night of frightening indicates
this effort was not successful.
Our data indicate that heron predation could be re-
duced substantially by firing pyrotechnics during re-
Table 2. Numbers and changes
in numbers
time [nights])
of black-crowned
and great
blue herons
Man at
the Watson
Lake Fish
Unit in Colorado
night-herons Great blue herons
with time Change
with time
Periods x SE Intercept Slope P x SE Intercept Slope P
1 66.7A" 5.8 87.0 -5.8 0.074 12.4A 1.7 7.2 1.5 0.153
1 23.8C 5.3 3.6 2.2 <0.0001 5.5C 1.3 0.7 0.5 0.0002
2 49.8B 3.2 56.8 -1.5 0.136 8.2BC 1.4 5.1 0.6 0.141
2 32.1C 3.1 36.8 -0.7 0.344 7.5BC 0.7 6.4 0.2 0.323
Posttreatment 2 27.7C 1.1 29.9 -0.5 0.195 9.8AB 0.7 11.4 -0.3 0.149
Numbers in a column
by the same
letter did not
(P> 0.05).
Deterring heron predation on fish * Andelt et al. 693
peated 14-night periods followed by 22 nights with-
out frightening. Assuming (as we observed) no
herons will capture fish during frightening and 9%
the BCNH
and 27%
of the GBH
would be present dur-
ing the 22-night periods following frightening, we es-
timate that the number of herons and heron preda-
tion on fish could be reduced to only 10%
and 27% GBH present during 61% of the season) of
pretreatment losses by firing pyrotechnics during 4
14-night periods from mid-April through early Sep-
tember. This amount of frightening (4 periods x 14
nights/period x 10 hrs/night) would cost $3,920
@$7/hr for labor and $840 for pyrotechnics (assum-
ing 4.0 pyrotechnics @$0.375 each [with firing cap]
are fired/hr). The costs of frightening ($4,760) and
fish lost (0.10 x 79,000 fish lost/yr during the 5 yrs be-
fore our study x $0.30/fish = $2,370) would be only
30% of the value of fish lost (79,000 x $0.30/fish =
$23,700) without frightening.
Bright rotating lights reduced the numbers of
BCNH and GBH on the western half of the rearing
unit by redistributing
some of the birds to the eastern
half where flashing lights were not operating. We
speculate the numbers of herons would not have
been reduced as much on the western half if flashing
lights had been placed on both sides because herons
traveled only a short distance to forage on the eastern
half. Although we hypothesized that the bright flash-
ing lights would inhibit the birds' ability to see and
capture fish, the lights did not affect the number of
attempted and successful captures of fish (Andelt, un-
publ. data).
Two Scarey Man mannikins reduced the numbers
and GBH at the rearing unit during the first
4 nights of the first treatment period, but the num-
bers of both species then increased substantially
through the eighteenth night of use. This trend, in
addition to the increased proportion of BCNH that re-
mained <2 minutes after a Scarey Man activated dur-
ing the first treatment period and the ineffectiveness
of Scarey Man during the second application, indi-
cated that herons quickly habituated to the devices.
We determined the effects of treatments by com-
paring counts of herons between pretreatment and
posttreatment or among pretreatment, treatment,
and posttreatment. Ideally, another rearing unit
should be used as a control to ascertain if changes in
numbers of herons were affected by treatments
rather than outside factors such as reproduction, in-
cubation, and feeding of young (Vos 1984). How-
ever, we believe our data adequately assessed the ef-
fects of treatments because: (1) our treatments lasted
<24 days, (2) numbers of herons remained fairly sta-
tic during 1993, 1 year after our study when we
tested a rather unsuccessful frightening device dur-
ing a short period (11 nights; Andelt and Hopper
1996), (3) numbers of herons generally were similar
during pretreatment periods of the 3 years of our
study and during 1993, (4) numbers of herons gener-
ally decreased during treatment compared to both
pretreatment and posttreatment periods, and (5) the
effects of pyrotechnics were similar during 1990 and
1991 although applied during late May through early
June 1990 versus late July 1991.
Management recommendations
Our data indicate that nylon-mesh barriers can be
used in fish-rearing
facilities to reduce GBH
on fish. However, the adaptability
of GBH
in choos-
ing alternate foraging sites suggests that barriers will
need to be placed at most if not all foraging sites to
greatly reduce predation. Bright rotating lights and
Man were ineffective in frightening BCNH and
GBH from the Watson Fish Rearing
We reduced numbers of BCNH 91%
and GBH
during a 22-night period following 14 nights of haz-
ing, but we were less effective when employing py-
rotechnics for a 7-night period. We believe fishery
managers can effectively reduce the costs of moder-
ate to substantial heron predation at fish-rearing
ities by employing pyrotechnics for 14-night periods
followed by periods without frightening.
Acknowledgments. We thank G. G. Schoonveld
and D. C. Smeltzer, Colorado Division of Wildlife, for
funding the study. D. C. Smeltzer provided data on
heron predation on fish and encouraged the evalua-
tion of pyrotechnics for deterring heron predation.
We thank M. M. Robinette for conducting most of the
field work during the second year and L. S. Butler for
summarizing data. K. P. Burnham provided sugges-
tions on statistical analyses. L. H. Carpenter provided
the night-vision device. B. A. Knopf and K. M. Giesen
reviewed the manuscript.
Literature cited
1996. Effectiveness of alarm-dis-
tress calls for frightening herons from a fish rearing facility.
Prog. Fish-Cult.
D. 1987. The effectiveness of attempts to reduce pre-
dation by fish-eating birds: a review. Biol. Conserv. 41:219-
K. F. 1939. The control of fish predators at hatcheries and
rearing stations. J. Wildl. Manage. 3:169-179.
NAGGIAR, M. 1974. Man vs. birds. Fla. Wildl. 27:2-5.
J. A., R. P. BROOKS,
1987. A survey of
wildlife depredation and control techniques at fish-rearing
cilities. Wildl. Soc. Bull. 15:386-394.
694 Wildlife
Society Bulletin 1997, 25(3):686-694
PARKHURST,J. A., R. P. BROOKS, AND D. E. ARNOLD. 1992. Assessment
of predation at trout hatcheries in central Pennsylvania. Wildl.
Soc. Bull. 20:411-419.
SALMON, T. P., AND F. S. CONTE. 1981. Control of bird damage at
aquaculture facilities. U.S. Fish and Wildl. Serv. Wildl. Manage.
Leafl. No. 475. llpp.
users guide, release 6.03 ed.
Inst., Inc., Cary, N.C. 1028pp.
F. W. 1946. An approximate distribution of estimates
of variance components. Biometrics Bull. 2:110-114.
AND R. E. STUITZ. 1978. Extent and
severity of avian predation at federal fish hatcheries in the
United States. Proc. Annu. Conf. Southeast. Assoc. Fish and
Wildl. Agencies 32:470-473.
A. R., AND K.
1989. Survey of Mississippi cat-
fish farmers on means, effort, and costs to repel fish-eating
birds from ponds. Proc. Eastern Wildl. Damage Control Conf.
A. R., J. F. GLAHN, J. O. KING, AND D. T. KING. 1995a. Im-
pact of great blue heron depredations on channel catfish farms.
J. World Aquacult. Soc. 26:194-199.
A. R., D. F. Morr, ANDJ.
O. KING. 1995b. Short-term ef-
fects of an inflatable effigy on cormorants at catfish farms.
Wildl. Soc. Bull. 23:73-77.
Vos, D. K. 1984. Response of breeding great blue herons to hu-
man disturbance in northcentral Colorado. M.S. Thesis, Col-
orado State Univ., Fort Collins. 65pp.
William (Bill) F. Andelt (photo) is an associate professor and ex-
tension wildlife specialist in the Department of Fishery and Wild-
life Biology at Colorado State University. He received his B.S.
(1974) and M.S. (1976) degrees in wildlife sciences from the Uni-
versity of Nebraska and his Ph.D. (1982) in zoology from Col-
orado State University. Bill's interests include research and exten-
sion related to managing human-wildlife conflicts and the ecol-
ogy of carnivores. Timothy (Tim) P. Woolley is a biologist with
the Wyoming Game and Fish Department. He received his B.S
(1992) in wildlife biology from Colorado State University and his
M.S. (1995) in zoology and physiology from the University of
Wyoming where he evaluated ground and fixed-wing aerial
pronghorn composition surveys. Stuart N. Hopper has nearly
completed a B.S. in wildlife biology at Colorado State University
and recently has worked on a Canada goose monitoring project
for the Colorado Division of Wildlife.
... [31,72,73]) and bird deterrence from zones of human-wildlife conflict (e.g. [74][75][76]). More recent studies use bird distributions to estimate the ecological impacts of ALAN on birds (e.g. ...
Full-text available
Background Artificial light is ubiquitous in the built environment with many known or suspected impacts on birds. Birds flying at night are known to aggregate around artificial light and collide with illuminated objects, which may result from attraction and/or disorientation. In other contexts, birds are repelled by light-based deterrents, including lasers and spotlights. Artificial light can also change birds’ perceptions of habitat quality, resulting in selection or avoidance of illuminated areas. Studies documenting aggregation, deterrence, and habitat selection are typically considered separate literature bodies, but they actually study a common set of populations, interventions/exposures, and responses. Our systematic map provides a comprehensive, searchable database of evidence of the effects of artificial light on bird movement and distribution, increasing both the quantity and diversity of studies that are accessible for further comparison and synthesis. We identify and describe the evidence available for four secondary questions relevant to conservation or management: aggregation/mortality at structures with artificial lights, evidence that light attracts and/or disorients birds, light-based deterrent efficacy, and the influence of continuous illumination on habitat selection. Methods Using the principles of systematic reviews and methods published in an earlier protocol, we conducted an extensive and interdisciplinary literature search. We searched multidisciplinary citation indices as well as databases and websites specific to conservation, pest management, transportation, and energy. In our map, we included all studies reporting eligible populations (birds), interventions/exposures (artificial light), and outcomes (movement through space, behaviour preceding movement, or distribution). We evaluated the quantity of available evidence based on meta-data fields related to study context, population traits, light source characteristics, and outcome variables. We used these meta-data to identify relevant evidence for each secondary question and describe aspects of our secondary questions that may support reviews (evidence clusters) and others that require more research (knowledge gaps). Review findings We manually screened 26,208 articles and coded meta-data for 490 eligible studies in a searchable database, organizing the literature to facilitate future reviews and evidence-based management. Much of the evidence was concentrated in particular locations (Northern hemisphere), taxonomic orders ( Passeriformes, Charadriiformes , and others), and light wavelengths (red and white). We identified 56 distinct response variables and organized them into 3 categories (behaviour, distribution, and avian community), showing the diversity in bird responses to light. Conclusions Our database can be used to answer the secondary questions we identified and other questions about the effects of artificial light on bird movement and resulting changes to distribution. There may be sufficient evidence for a review of the weather and lunar conditions associated with collisions, which could help identify nights when reduction of artificial light is most important. Further experiments should investigate whether specific types of light can reduce collisions by increasing the detectability of structures with artificial lights. The efficacy of lasers as deterrents could be evaluated through systematic review, though more studies are needed for UV/violet lasers. To reduce the impacts of outdoor lighting on birds, research should investigate how spectral composition of white light influences bird attraction, orientation, and habitat selection.
... birds away. This device was also found to be less effective than other deterrents in previous research 48 . Finally, the maximum concentration of methyl anthranilate (7.5%) specified on the product label sprayed on the ground did not appear to deter birds, which was surprising considering we applied this concentration for three consecutive days. ...
Full-text available
Toxic baiting of wild pigs ( Sus scrofa ) is a potential new tool for population control and damage reduction in the US. Field trials testing a prototype toxic bait (HOGGONE 2 containing 5% sodium nitrite [SN]), though, revealed that wild pigs spilled small particles of toxic bait outside of bait stations which subsequently created hazards for non-target species that consumed those particles, primarily passerine birds. To deter non-target birds from consuming particles of spilled bait, we tested four deterrents at mock bait sites (i.e., baited with bird seed) in north-central Colorado, USA during April–May 2020. We found a programable, inflatable deterrent device (scare dancer) reduced bird visitation by an average of 96%. Then, we evaluated the deterrent devices at SN-toxic bait sites in north-central Texas, USA during July 2020, where the devices were activated the morning following deployment of SN-toxic bait. Overall, we found 139 dead wild pigs at 10 bait sites following one night of toxic baiting, which represented an average of 91% reduction in wild pigs visiting bait sites. We found that deterrent devices were 100% effective at deterring birds from toxic bait sites. We found two dead non-target mice at bait sites without deterrent devices. We noted that deploying toxic bait in mid-summer rather than late-winter/early-spring reduced hazards to migrating birds because they were not present in our study area during July. We recommend using deterrent devices (i.e., novel, programmable, battery operated, continuous and erratic movement, and snapping sounds) to reduce hazards to non-target birds at SN-toxic bait sites. We further recommend deploying SN-toxic bait during seasons when migrating birds are not as abundant until further research demonstrates minimal risks to migrating birds.
... The results from our study appear similar to the only other experimental evaluations of inflatable effigy we could find, where Stickley et al. (1995) successfully used a similar human effigy to deter cormorants from consuming farmed catfish. The results from our trials also reflect the promising results reported by those evaluating similar frightening devices, including, for instance, Andelt et al. (1997);Beringer et al. (2003);VerCauteren et al. (2003) and Zarco-González and Monroy-Vilchis (2014). Although we found no strong evidence of habituation during the three consecutive trials of the inflatable effigy, this will need to be considered during longer-term trials. ...
Human-carnivore coexistence can be aided through non-lethal approaches that limit interaction between predators and livestock. Yet, investigations into effective deterrents, particularly in the Australian context with dingoes, are rare. We investigated two potential methods: an acoustic deterrent (series of gunshot noises), and an oversized inflatable human effigy that we dubbed ‘Fred-a-Scare’. The devices were deployed to determine whether they would deter captive dingoes (n=12), from accessing food. The acoustic deterrent did not appear to repel the dingoes during the first trial (11/12 accessing the food; the same as control). However, use of the effigy device was associated with a significant reduction in dingoes approaching, with only 25% (9/36) accessing food across all trials. On the third and final trial (which were repeated daily), 42% (5/12) of dingoes accessed food. Used in conjunction with other devices and methods, and at intervals that reduce the risk of habituation, the inflatable effigy could provide a valuable tool for deterring dingoes, and perhaps other species, from particular areas, even where food (or potential prey) is present. This has potential for use in human-dingo conflict hotspots, such as campgrounds and some small livestock enterprises, but field trials are required to evaluate the technique in these contexts and with free-ranging dingoes.
... As for the literature associated with bird aggregation and attraction to light, there are few generalizations with which to predict or evaluate the effect of light as bird deterrents. For example, a particular light-based treatment may effectively deter some bird species and not others [6,25], a deterrent that was initially effective may no longer induce a reaction after birds habituate [25], and small scale movements by birds as pests may simply relocate them to nearby areas [27,28]. The literature reflects a general consensus that some types of lasers and strobes can deter some bird species in some contexts [3,10,12], but no systematic comparison has been applied to understand when and why particular treatments are effective. ...
Full-text available
Background Anthropogenic light is known or suspected to exert profound effects on many taxa, including birds. Documentation of bird aggregation around artificial light at night, as well as observations of bird reactions to strobe lights and lasers, suggests that light may both attract and repel birds, although this assumption has yet to be tested. These effects may cause immediate changes to bird movement, habitat selection and settlement, and ultimately alter bird distribution at large spatial scales. Global increases in the extent of anthropogenic light contribute to interest by wildlife managers and the public in managing light to reduce harm to birds, but there are no evidence syntheses of the multiple ways light affects birds to guide this effort. Existing reviews usually emphasize either bird aggregation or deterrence and do so for a specific context, such as aggregation at communication towers and deterrence from airports. We outline a protocol for a systematic map that collects and organizes evidence from the many contexts in which anthropogenic light is reported to affect bird movement, habitat selection, or distribution. Our map will provide an objective synthesis of the evidence that identifies subtopics that may support systematic review and knowledge gaps that could direct future research questions. These products will substantially advance an understanding of both patterns and processes associated with the responses of birds to anthropogenic light. Methods The protocol describes the steps taken to ensure the search for evidence is comprehensive, transparent and replicable. We will find relevant studies in the grey and peer-reviewed literature using publication databases, Google Scholar, stakeholder suggestions, and organizational websites. We will select studies for inclusion in the map by identification of relevant: (i) population including any species of bird; (ii) intervention or exposure to anthropogenic light; and (iii) outcomes including changes in bird movement, habitat occupancy, population density, or distribution. We will extract and organize metadata into a systematic map that can support subsequent search by interested individuals. The quantity of evidence on particular topics will be characterized through heat maps and narrative syntheses, but subsequent work will be needed to evaluate evidence validity.
... Bioacoustic techniques have been implemented to deter birds in other instances, including landfills (Baxter 2000), airports (Blokpoel 1976), fish-rearing ponds (Andelt et al. 1997) and along highways (Conklin et al. 2009). Bioacoustic calls are thought to be the most effective audio deterrent for birds, because they invoke a natural fear response (Marsh et al. 1992). ...
Full-text available
Proceedings of the Nova Scotian Institute of Science. Wind energy has become one of the fastest-growing renewable electricity sources globally, and this trend is expected to continue. However, wind turbines cause avian mortality when birds collide with these structures. Although regulatory agencies in many jurisdictions require post-construction bird mortality monitoring at turbine sites, resulting mortality estimates are often imprecise and under-reported. This uncertainty is often attributed to searcher inefficiencies or scavenger losses. Furthermore, data regarding the effectiveness of active bird mortality mitigation at these facilities are also lacking. This pilot study assessed mitigation effectiveness of visual and audio deterrents, using predator owl deterrent models and bioacoustic alarm and predator calls deployed at a wind turbine facility in Nova Scotia, Canada. These deterrents did not deter birds from wind turbines in statistically significant ways, in comparison to control sites. Whilst results were inconclusive, it would be prudent to continue assessing mitigative options to minimize impacts on birds, considering the expected growth of the wind energy sector in Canada.
... recognition of the snake : Smith 1975;Matheus et al. 1996;Kullberg et al. 1998; recognition of a raptor : Walters 1990;Lind et al. 2003;Templeton et al. 2005;Schleidt et al. 2011;2D eye-like shapes;Jones 1980) or for applied reasons (e.g. Conover 1979;Hothem and DeHaven 1982;Cummings et al. 1986;Erickson et al. 1990;Marsh et al. 1992;Andelt et al. 1997;Harris and Davis 1998). For example, Smith (1977) suggested that naïve young great kiskadees (Pitangus sulphuratus) show spontaneous avoidance of the ring pattern of coral-snakes and of visual patterns reminiscent of them. ...
Full-text available
Stimuli such as visual representations of raptors, snakes, or humans are generally assumed to be universally fear-inducing in birds and considered as a product of evolutionary perceptual bias. Both naïve and experienced birds should thus react to such stimuli with fear reactions. However, studies on different species have shown the importance of experience in the development of these fear reactions. We hypothesized that the responses of adult European starlings to fear-inducing visual stimuli may differ according to experience. We compared the reactions of Hand-raised adults with no experience of predators to those of Wild-caught adults, with potentially extensive experience with predators. Three visual stimuli (i.e. human, raptor, snake) were broadcast to 17 birds as 2D pictures (displayed via a LCD screen) with different modalities of presentation: degree of proximity and with or without movement. The results reveal that the birds were particularly sensitive to proximity and movement, with more attention towards moving stimuli and more withdrawal for close stimuli. The human stimulus elicited attention in both the distant and moving modalities but, like the other stimuli, mostly withdrawal when it was close. Developmental experience appeared to influence the emotional level, as the Hand-raised birds reacted strongly to all stimuli and all modalities, contrarily to the WC birds which performed withdrawals almost only for close stimuli and attention to moving stimuli. Stimuli proximity and movement seemed, therefore, relevant features that elicited negative reactions in Wild-caught birds. The Hand-raised birds were equally attentive to both distant and moving stimuli. Thus the young birds showed no real discrimination. Early and later experiences may, therefore, influence birds’ reactions. Starlings may require experience with real threats to develop adaptive responses, i.e. limiting unnecessary loss of energy by fleeing in front of non-dangerous stimuli.
... Responses due to neophobia or initial confusion are likely to be particularly susceptible to habituation, such that after repeated, benign exposure, responses wane. This has been suggested for most of what have been labelled as 'frightening devices' that often feature a variety of loud sounds and/or bright lights as stimuli (Andelt et al. 1997;Conover 2002;Gilsdorf et al. 2002;Koehler et al. 1990;Smith et al. 2000b;Mason 2001). However, there are stimuli that are far less susceptible to habituation, often ascribed the term 'unconditioned stimuli' (Domjan 2010), because they tend to inherently produce responses even when following repeated exposure (exceptions being when, for example, satiation or exhaustion occur). ...
Existing methods of reducing livestock depredation are heavily biased towards lethal control. However, criticism regarding the efficacy of such practices is rising. In Australia, over 200 years of lethal control has done little to resolve the conflict between dingoes (Canis dingo) and livestock producers. That is, killing dingoes does not necessarily prevent livestock losses. Rather than continuing with lethal control programs, there is an opportunity to shift to more innovative, effective and ethical non-lethal measures of protecting livestock from attacks. Traditionally, buffer zones (areas surrounding livestock enterprises or national parks where attempts are made to eradicate all dingoes entering that zone) have been put in place as a means to limit conflict. Although seen as more strategic than indiscriminately baiting over large areas, targeting dingoes in buffer zones does not necessarily remove problem animals. In addition, dingoes from outside baited zones eventually fill any territorial voids created. In order to break this cycle, we propose amending the traditional approach, so that instead of killing dingoes in these sensitive zones, they are excluded from production areas or otherwise discouraged from interacting with livestock (what we term 'living buffer zones'). This can, in principle, be achieved through adoption of a suite of non-lethal management approaches, including aversive conditioning, which to-date has not been widely examined. In turn, resident dingoes conditioned to avoid livestock and/or livestock areas will maintain territories that largely exclude non-resident dingoes. Occasional ingress by transient dingoes will be met by the same exclusion and aversive strategies and are likely to quickly move on if harassed by resident dingoes. Such a strategy takes advantage of our ever-increasing knowledge of dingo biology and behaviour and leverages well established principles of animal learning. By funnelling funds currently spent on killing dingoes into experimental investigations of non-lethal approaches, we conclude that significantly more livestock will actually be saved.
... A common assessment of many fear-evoking or frightening devices is that they work for only short periods and responses are subject to habituation (Koehler et al. 1990;Andelt et al. 1997;Smith et al. 2000;Mason 2001;Conover 2002;Gilsdorf et al. 2002). Extremely loud noises or bright lights (e.g. ...
Full-text available
K'gari (Fraser Island) offers a rare opportunity for people to observe and encounter wild dingoes. Occasionally, however, such encounters can entail dingoes acting in a threatening or aggressive manner towards people, resulting in human injury and, in one tragic case, death.Asuite of approaches aimed at minimising the risk to human safety posed by dingoes have been implemented on the island, including fencing, island-wide warning signage, and regulations against feeding. Despite such measures, negative encounters continue, and in cases where dingoes are deemed to pose an unacceptable risk, they are usually destroyed. In searching for non-lethal management alternatives, attempts have been made to modify undesirable dingo behaviour through aversive conditioning, but results to date have either been mixed or largely disappointing. Here we review a wide array of research that has utilised aversive stimuli in an effort to modify and manage the behaviour of wild animals, with a particular focus on related predators such as coyotes and wolves. We identified eight major categories of experimental research: conditioned taste aversion/avoidance (CTA), electric fencing, fladry, chemical repellents, fear-evoking stimuli, physical repellents, aversive collars/devices and hard release procedures. We then outline each of these categories in more detail, complete with pertinent examples of successes and failures as well as advantages and disadvantages. We conclude that some approaches offer promise within three main areas of incident mitigation experimentation: dingo exclusion (e.g. electric fencing), personal protection (mild chemical irritant sprays, sturdy umbrellas) and remedial aversive conditioning (e.g. shock collars). Other approaches, such as CTA and sublethal projectiles are not recommended. Like any approach, aversive conditioning is not a panacea, but it does offer promise in filling gaps in current management and as an alternative to lethal control.
Full-text available
Weeds, pathogens, and animal pests are among the pests that pose a threat to the productivity of crops meant for human consumption. Bird-caused crop losses pose a serious and costly challenge for farmers. This work presents a survey on bird deterrent solutions for crop protection. It first introduces the related concepts. Then, it provides an extensive review and categorization of existing methods, techniques, and related studies. Further, their strengths and limitations are discussed. Based on this review, current gaps are identified, and strategies for future research are proposed.
Full-text available
We evaluated the effectiveness of using alarm-distress calls of the black-crowned night heron Nycticorax nycticorax and the great blue heron Ardea herodias for frightening these piscivorous bird species from a rearing unit for rainbow trout Oncorhynchus mykiss. Heron alarm–distress calls reduced black-crowned night heron numbers 48% during the 11-d treatment period, compared with the pretreatment period, but did not affect the number of great blue herons. The number of black-crowned night herons using the rearing unit and the proportion that remained immediately after the broadcast of alarm–distress calls increased from nights 1 through 11 of the treatment period, which indicates that this predator quickly habituated to the calls. We did not find any differences between the proportions of black-crowned night herons and great blue herons that remained on the rearing unit immediately after playing calls of the respective species. Alarm–distress calls were mostly ineffective in reducing heron predation on rainbow trout fingerlings.
Full-text available
Reduction of the damage caused by fish-eating birds requires accurate bird identification and some knowledge of avian biology and habits. The open-water areas and large concentrations of aquatic livestock at aquaculture facilities are natural attractants to many birds. Birds can have a significant economic impact on the culture of aquatic products including fish, shellfish, crustaceans, and other invertebrates. For our purpose here, we will refer to these birds as fish-eating birds, and the aquatic products as fish.
Surveyed 336 fish hatchery managers predominantly from the E USA, of which 85% responded, to assess aspects of hatchery depredation. Most facilities (63%) experienced losses to predators and managers identified 4 species of insects, 13 species of herpetofauna, 18 species of mammals, and 43 species of birds as predators. Control devices were used by 85% of the managers to limit depredation losses; top screens, enclosures, and shooting were rated as most effective in deterring predators and visual scare devices as least effective. Most managers (73%) could not place an economic value on their losses. -from Authors
The authors identified seven important predators at ten trout hatcheries and quantified their foraging behaviors. Most species devoted <28% of their total time in a hatchery to foraging for fish. As a group, predators fed on fish of variable lengths, but many factors determined specific lengths of fish taken. Most predators foraged equally at earthen and concrete raceways, but great blue herons Ardea herodias foraged only at earthen raceways. Collectively, common grackles Quiscalus quiscula captured and removed the most fish/day, but because mallards Anas platyrhynchos were present for more days/year, they had the greatest total impact. Secondary infection and injuries sustained during attacks by predators also may be important factors in fish loss. -from Authors
Eighty-seven percent of 281 Mississippi catfish farmers felt that fish-eating birds were enough of a problem to warrant harassment. Fanners estimated that they spent an average of 2.6 man-hours per day harassing birds at an average annual cost of $7400. Sixty percent of the farmers who harassed birds did so by driving around the ponds and firing at birds to repel them. Propane exploders and pyrotechnics were also used. Pyrotechnics were judged the most effective repellent technique. The cost of bird harassment according to these farmers is $2.1 million. The value of fish loss to cormorants alone is roughly estimated at $3.3 million. The total annual loss to birds is estimated at $5.4 million.
With the growth of channel caffish Ictalurus punetatus production in the Delta Region of Mississippi over the last 30 years have come concurrent depredation problems caused by great blue herons Ardea herodias. Biomass in stomachs from herons collected at catfish farms averaged 41% catfish, 38% sunfish Lepomis sp., 17% shad Dorosoma cepedianum, and 4% gambusia Gambusia sp.; whereas observations of herons foraging indicated that 45% of the prey taken were gambusia. Our observational data indicate that herons take an average of 12 10-cm catfish fingerlings daily. The diurnal density of foraging herons on catfish ponds averaged 0.17 herons/ha in 1990, which means that the average 127-ha farm supports approximately 22 herons. Nocturnal foraging, especially on dark nights, appears to be minimal. If our data are approximately correct, the average catfish farm could be losing $30/ha per yr to herons, assuming that this catfish fingerling mortality can be attributed solely to heron depredation.
The effectiveness of different types of attempts to control predation by fish-eating birds is reviewed in view of the increased abundance of both fish-farms and several fish-eating bird species, and the consequent increasing number of conflicts between birds and man and fish-farmers and bird protectors. There is, so far, no scientific basis to claim that removal of predators through relocation or killing (the method favoured most by farmers and fishermen) reduces bird abundance or increases fish yields. On the other hand, most bird deterrents (the alternative proposed by many bird protectors) have only a very short-term effect, if any, depending mainly on the availability of alternative resources. ‘Ecological’ measures have not yet been properly tested, but are unlikely to solve the problem. The only reasonable solution to prevent fish from being eaten by piscivorous birds is to make sites inaccessible through wiring and netting. The costs of such measures have to be carefully balanced against the expected reduction of fish losses.
Typescript (photocopy). Thesis (M.S.)--Colorado State University, 1984. Includes bibliographical references (leaves [56]-65).