Content uploaded by David A. Buehler
Author content
All content in this area was uploaded by David A. Buehler on Jan 05, 2016
Content may be subject to copyright.
Wiley and Wildlife Society
are collaborating with JSTOR to digitize, preserve and extend access to
The Journal of Wildlife
Management.
http://www.jstor.org
Effects of Human Activity on Bald Eagle Distribution on the Northern Chesapeake Bay
Author(s): David A. Buehler, Timothy J. Mersmann, James D. Fraser and Janis K. D. Seegar
Source:
The Journal of Wildlife Management,
Vol. 55, No. 2 (Apr., 1991), pp. 282-290
Published by: on behalf of the Wiley Wildlife Society
Stable URL: http://www.jstor.org/stable/3809151
Accessed: 05-01-2016 16:17 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
EFFECTS
OF
HUMAN ACTIVITY ON BALD
EAGLE
DISTRIBUTION
ON THE
NORTHERN
CHESAPEAKE
BAY
DAVID
A.
BUEHLER,
Department
of
Fisheries
and
Wildlife
Sciences,
Virginia
Polytechnic
Institute and
State
University,
Blacks-
burg,
VA
24061
TIMOTHY
J.
MERSMANN,'
Department
of
Fisheries and Wildlife
Sciences,
Virginia
Polytechnic
Institute
and State
University,
Blacksburg,
VA
24061
JAMES D.
FRASER,
Department
of
Fisheries
and
Wildlife
Sciences,
Virginia Polytechnic
Institute
and
State
University,
Blacks-
burg,
VA
24061
JANIS K. D.
SEEGAR,
Chemical
Research,
Development,
and
Engineering
Center,
U.S.
Army,
Aberdeen
Proving
Ground,
MD
21010-5423
Abstract:
We
determined
the
relationship
between
bald
eagle
(Haliaeetus
leucocephalus)
distribution
and
human
activity
on the
northern
Chesapeake
Bay
shoreline
during
1985-89.
Only
55
of
1,117
locations
of
radio-tagged eagles
(4.9%)
occurred
in
the
developed
land-cover
type
(>4
buildings/4
ha),
although
18.2%
of
potential
eagle
habitat was
developed
(x2
=
428.9,
4
df,
P
<
0.001).
Eagle
use of the
shoreline was
inversely
related
to
building
density
(x2
=
22.1,
P
<
0.001)
and
directly
related
to
the
development
set-back
distance
(x2
=
5.3,
P
=
0.02).
Few
eagles
used
shoreline
segments
with boats
or
pedestrians
nearby
(P
<
0.001).
Only
360 of
2,532
segments
(14.2%)
had
neither human
activity
nor
shoreline
development.
Eagle
flush distances
because of
approaching
boats were
greater
in
winter than in
summer
(Z
=
264.9
vs.
175.5
m,
respectively,
P
=
0.001),
but
were similar
for adult and
immature
eagles
(?
=
203.7 vs.
228.6
m,
respectively,
P
=
0.38).
Of
2,472
km
of
shoreline on the
northern
Chesapeake,
894
km
(36.2%)
appears
to be
too
developed
to be
suitable for
eagle
use,
and
an
additional 996 km
(40.3%)
had
buildings
within
500
m,
thereby reducing
eagle
use.
The
projected
increase in
developed
land in
Maryland
(74%)
and
Virginia
(80%)
from 1978
to
2020
is
likely
to determine the
future of the
bald
eagle population
on
the
northern
Chesapeake
Bay.
J.
WILDL.
MANAGE.
55(2):282-290
Eagle response
to
human
activity
varies with
food
supply,
disturbance
type,
perching
sub-
strate,
tide,
and overall
disturbance
level
(Knight
and
Knight
1984,
Fraser et al.
1985,
McGarigal
et
al.
1991).
Thus,
a
general
understanding
of
the
effects
of human
activities on
eagle
distri-
bution
requires
studies
of
eagle-human
inter-
actions
over
a
variety
of
habitats
and
ecological
conditions.
Such
an
understanding
is
crucial
to
the
long-term
conservation
of
this
species
in the
face of
human
development
in
shoreline habi-
tats
(Gray
et
al.
1988).
Moreover,
there
currently
are no
estimates
of
the minimum
building
den-
sity
that lowers
habitat
quality
for
bald
eagles.
Such
information could
help
land-use
planners
identify
habitat that
should
be
preserved
and
assess
the
impact
of
current and
proposed
de-
velopments
on bald
eagle populations.
We
studied the
relationship
of
bald
eagle
hab-
itat
use to human
development
and
other hu-
man
shoreline
activities
in
the
northern
Ches-
apeake
Bay.
Our
objectives
were
to
(1)
test the
hypothesis
that
eagle
distribution
was
indepen-
dent of
the
distribution of
humans
and human
developments,
(2)
examine
the effects
of
differ-
ent levels
of
building density
on
eagle
use of
the
shoreline,
(3)
evaluate
the
U.S.
Geological
Sur-
vey
(USGS)
Land
Use and Land
Cover
database
as
an
indicator
of
eagle
habitat
quality,
and
(4)
compare
flush distances
of
eagles
in the
Ches-
apeake
region
with
flush distances
reported
else-
where.
The U.S.
Army
Chemical
Research,
Devel-
opment,
and
Engineering
Center
funded
this
project.
A. B.
Jones
assisted
with
ARC/INFO
and
Advanced
Revelation
analyses.
We
thank
R. L.
Kirkpatrick,
D. F.
Stauffer,
D.
J.
Orth,
R.
L.
Knight,
E. P.
Smith,
and
an
anonymous
ref-
eree
for
reviewing
earlier
versions of the
manu-
script.
Other individuals
greatly
assisted
with
this research
in
a
variety
of
ways
and have
been
acknowledged
in
Buehler
et al.
(1991a).
STUDY
AREA
The
study
area extended
along
the
Chesa-
peake
Bay
from
the
Bay
Bridge
at
Annapolis,
Maryland
to
the
Conowingo
Dam on the
Sus-
quehanna
River,
encompassing
3,426
km2
and
including
2,472
km
of
bay,
river,
and
creek
shoreline
(Fig.
1).
Buehler
et
al.
(1991a)
de-
scribed the
study
area
in
more detail.
I
Present address:
U.S. Forest
Service,
Homochitto
National
Forest,
Gloster,
MS
39638.
282
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
J.
Wildl.
Manage.
55(2):1991
BALD
EAGLE-HUMAN
INTERACTIONS
*
Buehler et
al.
283
Of
the
2,199
km2
of
terrestrial habitat
on
the
study
area,
400
km2
(18.2%)
were classed as
developed
in
the
USGS
Land Use
and
Land
Cover data
base
(Anderson
et al.
1976).
Devel-
opment
categories
included residential
(201
km2,
14.0%),
commercial
(86
km2,
6.0%),
industrial
(43
km2,
3.0%),
transportation,
communication,
and utilities
(36
km2,
2.5%),
and
mixed and mis-
cellaneous
developed categories
(34
km2,
2.4%)
(Fig.
1).
Of
the
2,472
km
of
study
area
shoreline,
682
km
(27.6%)
were classed as
USGS-devel-
oped.
METHODS
Bald
Eagle
Distribution
Radio
Telemetry.-We
monitored
distribu-
tion
and
movements
of
nonbreeding
eagles
with
radio
telemetry. Eagle
capture, radiotagging,
and
telemetry
techniques
have been described
by
Buehler et al.
(1991a).
Shoreline
Surveys.-We
surveyed
the same
614
km
of
contiguous
shoreline
monthly
from
fixed-wing
aircraft. The
survey
route
(16.3%
USGS-developed)
under-represented
the devel-
oped
area on
the northern
Chesapeake
(27.6%
USGS-developed);
we excluded
the
intensively
developed
Baltimore area
because of federal
flight
rules.
Excluding
this area made
it difficult
to detect whether
eagles
were
avoiding
human
activity
because it excluded the
area with the
greatest
development
and the least
eagle
use.
However,
we never located a
radio-tagged eagle
there
during
the 3
years
of
the
study,
and no
nest sites were located
in the area. We also
ex-
cluded the Chester
River
drainage
because
of
time constraints.
We
surveyed
the
Eastern
Shore
on
Saturdays
and
Aberdeen
Proving
Ground
and
the
Susquehanna
River areas on
Sundays.
Sur-
veys began
about 30
minutes after sunrise
and
were flown
along
the shoreline
at 150
km
per
hour,
50 m above
the
water and
50
m
offshore,
for
about 2.5 hours duration.
We used
the same
pilot
and an
experienced
front
seat observer to
spot eagles,
pedestrians,
and
boats;
a
back seat
observer
recorded locations on
7.5-minute USGS
topographic
maps.
Human
Activity
and Shoreline
Development
We used
the USGS
Land Use and
Land
Cover
data base
(Anderson
et al.
1976)
to
identify
land-
cover
type
(developed,
farmland, forest,
wet-
land, other)
for
every
location
where
we saw
a
SUSQUEHANNA
CONOWINGO
DAM
RIVER
all,f
ABERDEEN
"
PROVING
GROUND
/ .
, EASTERN
S",, ,
?
SH1ORE
BALTIMORE
-,"
Lj
N
10 KM
MD
BAY
BRIDGE
S
:STUDY
AREA
USGS-DEVELOPED
I
AREAS
i
VA
Fig.
1.
Developed
areas
of the northern
Chesapeake
Bay,
Maryland,
based
on the U.S.
Geological
Survey
Land Use and
Land Cover
data base
(Anderson
et al.
1976).
radio-tagged
eagle.
In
this data
base,
land was
classed as
developed
if there were
>4
buildings
within
the minimum-sized cell
(4 ha).
To de-
termine
if
eagles
used land-cover
types
in
pro-
portion
to their abundance
on the
study
area,
we
compared
the
frequency
of
eagle
locations
in
each
cover
type
with the
proportion
of the
entire
study
area
composed
of that cover
type.
To evaluate
the
utility
of the USGS devel-
opment
classification for
delimiting
suitable ea-
gle
habitat,
we
located all
buildings
within
500
m
of
the
shoreline
survey
route on
1:12,000
color
aerial
photos
taken in 1985. We divided the
survey
route shoreline into
250-m
segments
with
Advanced Revelation
computer
software and
classed each
segment
as
used
by
eagles
if we
observed
an
eagle
within 50 m
of
that
segment
during any
of
the 36 aerial
surveys.
We esti-
mated the distance that
eagles
responded
to
hu-
man
activity
as 500
m,
based on our
maximum
flush
distance
(475 m)
and
the maximum
agi-
tation
response
distance
(486 m)
observed
by
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
284
BALD EAGLE-HUMAN
INTERACTIONS
*
Buehler et al.
J.
Wildl.
Manage.
55(2):1991
Table
1.
Shoreline
(km),
radio-tagged eagle
locations
(no.
locations/shoreline
km),
buildings (no./shoreline
km),
and
development
by geographic
area
(%),
northern
Chesapeake
Bay,
Maryland,
1985-88.
Susquehanna
Aberdeen
River
valley
Proving
Ground Eastern Shore
Baltimorea
Totala
Shoreline km
78
483
1,212
699
2,472
Eagle
locations/shoreline
km
2.9
2.8
0.7
0.0
1.0
Buildings/shoreline
km
4.3
3.6
11.9
8.3
%
developed/shoreline
kmb
10.9
19.1
7.8
69.1
27.6
a
The number
of
buildings
on
the
Baltimore
shoreline was
not
counted.
b
Developed
land
use was
determined
based on
the
U.S.
Geological
Survey
Land Use
and Land
Cover
data
base
(Anderson
et
al.
1976).
McGarigal
et al.
(1991).
A
segment
was
classed
as used
by
humans if
pedestrians
or
boats
were
observed
within 500
m
of that
segment
on
any
survey.
Developed
segments
for the aerial
photo
analysis
were
those
with
?1
building
within
500
m.
We
used
Chi-square
analyses
to test the
hy-
pothesis
that
eagle
use
of
shoreline
segments
was
independent
of the
presence
of
buildings
within
500
m of the shoreline.
We tested the
hypothesis
that
eagle
and hu-
man occurrences
on shoreline
segments
were
independent
of
one another
by
treating
each
individual
survey
as a
sample.
We did
this be-
cause we
thought
that
repeated
observations of
individual
segments
were not
independent
and
because
we wanted
to
allow
eagle
use to be
temporally paired
with
the observed
human ac-
tivity
in the
analysis.
For each
survey
we cal-
culated the
expected
number
of
observations
of
eagles
and humans on the same
250-m
shoreline
segment
under the model of
independence
as
SeSh/S,,
where
Se
=
the number of shoreline
segments
with
eagles,
Sh
=
the number
of shore-
line
segments
with
humans,
and
S,
=
the total
number of
shoreline
segments.
We used
a Wil-
coxon
signed-rank
test
to test for
differences
be-
tween the observed
number of shoreline
seg-
ments
with
eagles
and
humans
present
simultaneously
versus the
expected
number of
such
segments.
After
detecting significant
dif-
ferences with the Wilcoxon
test,
we conducted
Chi-square
tests of
independence
between
eagle
presence
and human
presence
to
determine
which
surveys
contributed
most to the
signifi-
cance of the overall
test.
We calculated
building
density
in
an area that
included
all
points
within 500
m
of
each
250-m
shoreline
segment.
We
also calculated
the dis-
tance from the
shoreline
to the closest
building,
defined as the set-back distance.
We
used the
LOGIST
procedure
(SAS
Inst.,
Inc.
1986)
to
de-
velop
a
logistic regression
model of the
depen-
dent
variable
eagle
use
(eagle
presence)
and the
independent
variables
shoreline
building
den-
sity
and
set-back
distance.
We
used an a-level
of 0.05 for
accepting
a variable in
the model.
We
also calculated
building density
in
an area
that
included all
points
within
100 m
of the
shoreline
to
compare
with
the
USGS land-cover
classification. This
area
on
a
250-m
shoreline
segment
(4.07
ha)
was
similar in size
to the
min-
imum
cell size
(4
ha)
used
in
the USGS
system.
From
our aerial
photos
we
classed shoreline
seg-
ments
based on
building density
categories
of
no
buildings,
0.01-0.99
buildings
per
ha,
and
S1
building
per
ha. We
used a
Chi-square
anal-
ysis
to test
whether the
USGS
system
and
our
aerial
photo
classification were
independent.
During
1987-89,
we
estimated
eagle flushing
distance
by approaching
perched
eagles
by
boat
(-3
km/hr)
from a
distance
>500
m
at an
angle
perpendicular
to the
shoreline until
they
flew.
We
measured the
distance from the
observer to
the
eagle perch
with
a line
marked
in
meters.
Eagles
were flushed
during
all times of
the
day,
throughout
the
study
area,
during
summer and
winter. We
spread
our
flushing activity
out
geo-
graphically
and over 2
years
to avoid
individuals
becoming
habituated
or
sensitized to our
ap-
proach.
RESULTS
Shoreline
Development
and Human
Activity
We counted
5,120
buildings
within 500
m
of
the
surveyed
shoreline. Aberdeen
Proving
Ground and the
Susquehanna
River
had
similar
development
intensity,
whereas
the Eastern
Shore
had more
buildings
along
the shoreline
(Table 1).
Of the
2,532
survey
route
segments,
1,673
(66.8%)
had
buildings
within 500
m of
the
shore,
with an
average
density
of 0.09
+
0.003
(SE)
buildings per
ha. Within
100 in
of
the
shoreline,
building density
was
significantly
greater
(:
=
1.24
?
0.05
buildings/ha,
P
<
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
J.
Wildl.
Manage.
55(2):1991
BALD
EAGLE-HUMAN
INTERACTIONS
*
Buehler
et
al. 285
Table
2.
Pedestrians and boats
(no./shoreline
km)
by
season and
geographic
area,
northern
Chesapeake
Bay, Maryland,
November
1985-August
1988.
Susquehanna
Aberdeen
River
valley Proving
Ground Eastern Shore Totala
n
SE SE
f
SE f
SE
Pedestriansb
Fall
7 0.23 0.07 0.05 0.02 0.18 0.04 0.13A 0.02
Winter 9
0.04 0.02
0.01
0.00 0.08 0.03
0.04B
0.02
Spring
9
1.16
0.32 0.08
0.02 0.14 0.04
0.17AC
0.04
Summer
9
0.96
0.18 0.10 0.02 0.29 0.08
0.24C
0.06
Totale
34 0.62
0.13 0.06 0.01 0.17
0.03
0.15 0.02
Boatsd
Fall
7
0.28
0.10 0.11 0.04 0.12 0.04 0.12A 0.04
Winter 9 0.00
0.00 0.00 0.00 0.01 0.00
0.00B
0.00
Spring
9 0.57
0.18
0.07
0.02 0.08
0.03
0.10A
0.03
Summer
9 0.76 0.13
0.21 0.06 0.46 0.13 0.38C 0.09
Totale
34 0.41 0.08
0.10 0.02 0.17 0.05
0.15 0.04
a
Values
followed
by
the same
capital
letter within
pedestrian
or boat
categories
are not different
(multiple
comparisons
with
pair-wise
Friedman's
tests,
P
>
0.05).
b
No.
pedestrians
per
shoreline
km differed
by geographic
sub-unit
and
season
(x2
approximations
of
Friedman's
test,
x2
=
34.28,
2
df,
P
<
0.01;
and
X2
=
32.32,
3
df,
P
<
0.01,
respectively).
c
Values underscored
by
the same horizontal
line within
pedestrian
or boat
categories
are not different
(multiple
comparisons
with
pair-wise
Friedman's
tests,
P
>
0.05).
d
No. boats
per
shoreline
km differed
by
geographic
sub-unit
and season
(x2
approximations
of
Friedman's
test,
X2
=
15.49,
2
df,
P
<
0.01;
and
x2
=
46.11,
3
df,
P <
0.01,
respectively).
0.001);
245
of the
developed segments
had
>1
building
per
ha.
Of
the
segments
with
__1
building
per
ha
within 100
m in the
aerial
photo
analysis,
143
(58.4%)
were classed
by
the USGS
as
developed.
Only
89
of 455
segments
(19.6%)
with 0.01-0.99
buildings
per
ha were classed
as
developed,
whereas
181
of
1,832
segments
(9.9%)
with
no
buildings
within 100
m were classed
as
devel-
oped by
the
USGS
(P
<
0.001).
We observed
0.15
pedestrians
per
km
of
shoreline
surveyed
(Table 2).
Pedestrian
activity
was
greatest
in summer
and least
in winter
(P
=
0.001),
but
spring
and
fall,
and
spring
and
summer
pedestrian
activity
did not differ
(P
=
0.08
and
0.83,
respectively).
Pedestrian
activity
was
greatest
along
the
Susquehanna
River,
in-
termediate
on
the Eastern
Shore,
and
lowest
on
Aberdeen
Proving
Ground
(P
<
0.05).
We observed
significantly
more
boats
per
shoreline
km in summer
than
in winter
(P
<
0.001)
(Table 2).
Boat
activity
in
spring
and
fall
was intermediate
and did
not differ
(P
=
0.90).
Boat
activity
was
greatest
on
the
Susquehanna
River
(P
<
0.05),
but did
not differ
on the
East-
ern
Shore
and
Aberdeen
Proving
Ground
(P
=
0.59).
We observed
pedestrians
within 500
m
of
1,425
shoreline
segments
(56.3%)
and
saw boats
within 500
m
of
1,748
shoreline
segments
(69.0%).
Segments
were used
inconsistently,
however.
We
observed
pedestrians during
>4
of 34
surveys
on
only
338 of
1,425
segments
(23.7%),
and we observed
boats
during
>4
of
34
surveys
at
only
701
of
1,748
segments
(40.1%).
We observed
pedestrians
close to the shoreline
(f
=
32.4
?
1.0
m),
whereas boats
were used
farther
from shore
(G
=
163.5
+
2.0
m)
(Fig.
2).
Pedestrians
and boats were observed
within
500
m of
developed
segments
more often than
would be
expected
if
the distributions
of
pe-
destrians and boats
were
independent
of
build-
ings
(P
<
0.001) (Table 3).
We observed
pe-
destrians, boats,
or
buildings
on
2,172
of
2,532
segments
(85.8%)
monitored.
We observed all
3
forms of
human
activity
on 983
of
2,532
seg-
ments
(38.8%).
Human-Eagle
Interactions
Radio-Tagged
Eagles.--Only
55 of
1,117
(4.9%)
locations of
radio-tagged
eagles
occurred
in
USGS-developed
land-cover
blocks,
although
18.2% of
the
study
area
was
developed,
a
sig-
nificant difference
(P
<
0.001) (Table
4).
Most
eagle
locations
in
human-developed
land-cover
types
(80.4%)
occurred on
Aberdeen
Proving
Ground.
Actual
human
activity
in most
of these
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
286
BALD EAGLE-HUMAN
INTERACTIONS
*
Buehler
et al.
J.
Wildl.
Manage.
55(2):1991
:0
100-
0
8
80
Eagles
N
w
3,418
ME
Pedestrians
N
a
2,999
Co
60
W
Boats
N
*
2,949
.0
O
-
Buildings
N
*
5,071
40
0
CL
o
25
75 125
175
225
275 325
375
425
475
Distance to
Shoreline
Midpoint
(m)
Fig.
2. Distance from
the shoreline
(m)
of bald
eagles,
pedestrians,
boats,
and
buildings during
aerial
surveys
of the northern
Chesapeake Bay, Maryland,
1985-88.
areas was
low, however,
and there
were few
buildings.
Only
11
of
1,117
eagle
locations
(0.9%)
occurred
in
nonmilitary-developed
areas,
al-
though
these
areas accounted
for 14.8%
of all
nonaquatic
land cover.
We never
located
radio-tagged eagles
in the
highly
developed
Baltimore
area,
despite
its 699
km
of shoreline
(Table
1).
We located
radio-
tagged
eagles
with similar
frequency
per
shore-
line
km
on
Aberdeen
Proving
Ground
and the
Susquehanna
River,
whereas
we obtained
about
25% as
many
locations on
the Eastern Shore
(Table
1).
Shoreline
Surveys.-We
observed
eagles
less
often
on shoreline
segments
with
buildings
than
would
be
expected
if
eagles
selected
segments
independent
of
development
(P
<
0.001) (Table
3).
The
magnitude
of the
effect
was
greatest
in
summer
(x2
=
76.1)
and least
in the fall
(x2
=
13.6).
Eagle
use of
shoreline
segments
was
in-
versely
related
to
the
density
of
shoreline
de-
velopment
(logistic
regression,
#-coefficient
=
-4.57,
SE
=
0.97,
x2
=
22.1,
P
<
0.001,
R
=
-0.11),
and
directly
related
to the
development
set-back
distance
(P-coefficient
=
0.001,
SE
=
0.001,
x2
=
5.3,
P
=
0.02,
R
=
0.04).
We observed
eagles
on
94
of 455
developed
shoreline
seg-
ments
(20.8%)
that had
0.01-0.99
buildings
per
ha
within 100
m,
but
only
17 of 245
segments
(6.9%)
that had
?1
building
per
ha; whereas,
we observed
eagles
on
565 of
1,832
segments
(30.8%)
with no
buildings
within
100
m
(x2
=
73.4,
2
df,
P
<
0.001).
We located
eagles
less often
than
expected
on
shoreline
segments
with
pedestrians
or
boats
within 500
m than on
segments
without
such
activities
(P
<
0.001) (Table
5).
Eagle
use of
pedestrian-used
segments
was lower
than
ex-
pected
(P
<
0.05)
on individual
surveys
in
May-
September
1987
and
August
1986 when
pedes-
trian
activity
was
greatest.
Eagle
use of
segments
with boat
activity
was lower
than
expected
(P
<
0.05)
on individual
surveys
in
July-August
1986,
April-September
1987,
and
May
1988
when boat
activity
was
greatest.
We seldom
ob-
served
eagles
within 500
m of human
activity
(f
=
1.0
?
0.3
segment/survey
for
eagles
and
pedestrians
and
t
=
2.2
?
0.6
segments/survey
for
eagles
and
boats).
Eagle
Flush
Distances
Eagles
flushed at the
approach
of
a boat at
40-475
m
(9
=
215
?
12
m)
(Table
6).
Flush
distances
caused
by
boats were
significantly
greater
in winter
than
in
summer
(P
=
0.001).
Adults and
immatures
flushed at
similar dis-
tances
(P
=
0.38).
Eagles
on
the Eastern Shore
flushed at
significantly
greater
distances
than
those
flushed
on Aberdeen
Proving
Ground
(P
=
0.05).
DISCUSSION
Bald
eagles
on
the northern
Chesapeake Bay
rarely
used
developed
areas or
areas
frequented
by
people
in boats or on
foot. We observed
no
eagle
use of
the Baltimore
area,
apparently
be-
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
Table
3.
Developed
and
undeveloped
250-m
shoreline
segments
with
pedestrians,
boats,
and bald
eagles
(no.
segments
in
class,
no.
expected
under the
hypothesis
of
independence
and associated
cell
x2),
northern
Chesapeake
Bay, Maryland,
November
1985-August
1988.
Segments
were classed
as
having
pedestrians,
boats,
or
eagles
present
if
at least
1
pedestrian
or boat was observed
within
500
m or
if
at least
1
eagle
was observed
within
50
m of
a
segment
on
at
least
1 of
34
aerial
surveys.
Seasona
Fall
Winter
Spring
Summer
All
yr
Obs
Exp
Cell
X2
Obs
Exp
Cell
x9
Obs
Exp
Cell
x2
Obs
Exp
Cell
x2
Obs
Exp
Cell
x2
Pedestrians
absentb
Undeveloped
723
573.3
39.06 765
671.1 13.15
749
614.7
29.33
767 558.1
78.21 602
375.6
136.53
Developed
967
1,116.7
20.06
1,213 1,306.9
6.75
1,063 1,197.3
15.06 878
1,086.9
40.16 505
731.4 70.10
Pedestrians
presentb
Undeveloped
136
285.7 78.40
94 187.9 46.96
110 244.3 73.80
92 300.9 145.05 257 483.4
106.06
Developed
706 556.3
40.26 460 366.1 24.11 610 475.7 37.89 795 586.1 74.47 1,168
941.6 54.46
Boats absentb
Undeveloped
566 526.2
3.01 852 836.9
0.27 528
482.1 4.37
450
354.5 25.71 367 266.0
38.37
Developed
985
1,024.8
1.55
1,615
1,630.1
0.14 893
938.9 2.25 595 690.5
13.20 417
518.0
19.70
Boats
presentb
Undeveloped
293 332.8
4.76 7 22.1 10.27 331 376.9 5.59
409
504.5 18.07 492 593.0
17.21
Developed
688 648.2 2.45 58 42.9 5.28 780
734.1
2.87 1,078 982.5
9.28 1,256 1,155.0 8.84
Eagles
absentb
Undeveloped
753
778.6
0.84
679
742.0 5.34
749 794.9 2.65
666 738.2
7.07 547
629.7
10.85
Developed 1,542
1,516.4
0.43
1,508 1,445.0
2.74
1,594 1,548.1
1.36
1,510 1,437.8
3.63
1,309
1,226.3
5.57
Eagles presentb
Undeveloped
106 80.4
8.15
180 117.0 33.86
110 64.1 32.83
193 120.8
43.19
312
229.3
29.79
Developed
131
156.6
4.18
165 228.0
17.39
79
124.9 16.86
163
235.2
22.18 364 446.7
15.30
a
Expected
and
cell
x2
values
were
based
on
individual
x2
analyses
conducted
in
each
season.
b
The
number
of
segments
with
pedestrians,
boats,
or
eagles
present
versus
absent
differed
for
developed
versus
undeveloped segments
over
all
seasons
and the entire
year
(x2
test,
P
<
0.001).
UtT
Cn
cs.
zD
HD
C)
z
00
-1
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
288 BALD
EAGLE-HUMAN
INTERACTIONS
*
Buehler
et al.
J.
Wildl.
Manage.
55(2):1991
.2
CL
t
C
E
0
0
o
a,
2
cc
t0
to
a4,
V)
"0
C
moo
?-
C
X
4:)
coa
U)
c
"O
~0c
0
r-
(D
4
C-
O
Ef
cc
I'
0aa
-J
a,c
.o0
4)n
Cr-
cE4
zo
ca
tc
U)
ca
*-,
.
o
9I
•cq
,oc~
e)D
o
ci
0
(0
0m
U-
"x
0
CIO
Z
li
1-4
w"
c
v
(0
CCO
-o
c
- O
t-
0
C)
1z)
OOI
0
CIV
0 0
r
UC
Cd
S-CU
So
Q
r,
0,
v
Ow
~0
C-1
N
(U
0
-o
To
0
C0
0
Cd
Cd
cause of the intense
development
of
nearly
70%
of the shoreline. Baltimore was much more de-
veloped
than the
Susquehanna
River,
Aberdeen
Proving
Ground,
and
Eastern Shore
areas,
all
of
which
had substantial
eagle
use.
Similarly, eagles
did
not use
much of
the
shoreline
in the
other 3 areas because
of
devel-
opment
and
human
activity.
That
shoreline
sur-
vey
segments
with
_1
building per
ha were
almost
never
used is
consistent
with
the lack
of
radio-tagged eagle
use of
the
sections considered
developed
(>4
buildings/4
ha)
by
the
USGS.
Segments
with
0.01-0.99
building
per
ha within
100
m
were used
by
eagles,
but
with
signifi-
cantly
less
frequency
than
segments
without
buildings
within 100
m.
Radio-tagged
eagles
used
areas
the USGS
identified as forested
or
in
wetland
land-cover
types
in
excess of
availability,
whereas
eagle
use
of
developed
and
agricultural
land-cover
types
was much
less than
expected
based
on
avail-
ability.
Thus,
this
system
is useful as
a first
ap-
proach
to
locating
an
area's
potential eagle
hab-
itat,
because
it can be used to
predict
quickly
which areas
might
be used
by
eagles.
Assuming
the entire Baltimore
area and all
other
USGS-developed
lands are
too
developed
for
eagle
use,
894
km of northern
Chesapeake
shoreline
(36.2%)
appear
unsuitable as
eagle
habitat.
Furthermore,
if the
survey
route is
rep-
resentative of
the northern
Chesapeake
exclud-
ing
the Baltimore
area,
an
additional 996
km of
shoreline
(40.3%)
are
substantially
degraded
habitat
because
buildings
are within
500
m.
Thus,
more than
76% of
the shoreline has limited
po-
tential as
eagle
habitat.
Also,
up
to an
additional
10%
of
the
shoreline was
unsuitable
for
eagle
use at
any given
time because of
boat
or
pe-
destrian traffic.
Eagle
sensitivity
to
human
activity,
as
mea-
sured
by
flush
distances,
was
similar to
values
reported
elsewhere. Our summer
flush distances
(i
=
176
m)
were
similar to
distances
reported
for the
Columbia
River,
Oregon-Washington
(t
=
197 m
[McGarigal
et al.
1991]).
Our
winter
flush
distances
(z
=
265
m)
were
greater
than
those
reported
for
canoe-induced flushes
in
Washington
(f
=
152
m
[Knight
and
Knight
1984]).
We detected
no difference in
response
by
age,
similar to
Knight
and
Knight
(1984),
but
unlike
Stalmaster and Newman
(1978).
The
mechanism
causing eagles
to
avoid areas
inhabited
by people
remains
unknown.
People
may
directly
flush
eagles
from
developed
shore-
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
J.
Wildl.
Manage.
55(2):1991
BALD
EAGLE-HUMAN
INTERACTIONS
*
Buehler et al.
289
Table 5. Observed and
expected
number
of
shoreline
segments
in
which
bald
eagles
and
pedestrians,
and
bald
eagles
and
boats were
present
during
aerial
shoreline
surveys,
northern
Chesapeake
Bay,
Maryland,
1985-88.
Eagles
and
pedestriansa
Eagles
and
boatsa
1985-86
1986-87 1987-88 1985-86
1986-87
1987-88
Month
Obs
Exp
Obs
Exp
Obs
Exp
Obs
Exp
Obs
Exp
Obs
Exp
Sepb
0.0 3.0 2.0
6.8c
6.0 9.0 6.0
14.Oc
Octb
0.0
1.8
1.0 4.9
0.0 3.2 4.0 6.8
Nov
1.0
0.6
3.0
7.6
1.0 2.2
0.0 0.6 0.0 3.2 0.0 0.4
Dec
0.0
0.0
0.0
5.9c
5.0 6.2
0.0
0.0 0.0 1.0 0.0
0.3
Jan
0.0 0.7 0.0 0.5 0.0 0.3 0.0
0.0
0.0
0.0 0.0
0.0
Feb 0.0
0.4
0.0 0.6
4.0 1.5c 0.0 0.1 0.0 0.0 0.0 0.6
Mar 0.0 1.5
0.0 1.3
1.0 2.7
0.0
0.9 0.0 0.6 0.0 0.4
Apr
0.0
0.8
2.0 2.8
0.0 0.8
0.0
1.7 0.0
3.5c
1.0
0.8
May
0.0 2.1
0.0 3.5c 0.0
2.1 1.0 3.9 0.0 6.9c 0.0
3.2c
Jun
2.0 4.3
0.0 13.6c 0.0
1.8
5.0
9.4 10.0
18.1c
1.0
5.1
Jul
6.0 9.7
1.0
6.6c
1.0
2.4
17.0
27.2e 5.0
12.4c
3.0
4.4
Aug
3.0 10.3t
0.0
6.7c
0.0 3.0
8.0 25.7c
8.0
15.6c 9.0
9.7
a
Observed
number
of
segments
with
eagles
and
pedestrians,
and
eagles
and
boats
present
was
significantly
less than
expected
(Wilcoxon
signed-
rank
tests,
P
<
0.001).
b
Human
activity
was not recorded
during September
and
October
1985
surveys.
c
Eagle
and
pedestrian
presence
on
segments,
and
eagle
and
boat
presence
on
segments
were not
independent
(x2
tests of
independence,
P
<
0.05).
line,
or
eagles
may
avoid
shoreline
with
people
present. Flushing
by
people
occurred
infre-
quently
on
the Columbia
River
(McGarigal
et
al.
1991).
In that
study,
eagles
avoided
flying
into
areas
where boats
were
present.
Eagles
on
the northern
Chesapeake
also
avoided
devel-
oped
shoreline
in winter
when there
were
low
levels of
human
activity, suggesting
that
eagles
may
respond
to structures
themselves
or
to hab-
itat alterations
such
as the removal
of
trees,
which
may
be
associated
with
development.
The
greater
flush distances
we
observed
in
winter could
be
due
to
seasonal
changes
in in-
dividual tolerance.
Eagles
might
become
ac-
customed
to
humans
in
summer
when
human
activity
is
common
and
less
avoidable,
but
eagles
in
winter
might
be
more
sensitive
because
they
are
less
frequently
exposed
to
people.
The
great-
er flush
distances
observed
on Aberdeen
Proving
Ground,
where
human
activity
was less
common
than the
Eastern
Shore,
are
consistent
with
this
interpretation.
These
results
suggest
that
eagle
habituation
to human
activity
may
have
oc-
curred,
which
was similar
to observed
differ-
ences in
eagle
response
to
experimental
flushing
on
rivers
with
high
and low
levels
of
human
activity
in
Washington
(Knight
and
Knight
1984).
Alternatively,
seasonal
differences
in flush dis-
tances
might
be
due
to
different
responses
in
different
eagle
populations.
Eagles
from
the
southeastern
United
States
are
present
on the
Chesapeake
in
summer,
and
eagles
from
the
northeastern
United States
and
Canada
are
pres-
ent in
the winter
(Buehler
1991b).
MANAGEMENT
IMPLICATIONS
Our
study
and others
have shown that
eagles
avoid
developed
and
human-used
habitats.
Yet
the more isolated
forested shorelines
used
by
eagles
are
rapidly
being
converted
to
housing,
recreational
facilities,
and
industrial
plants
(Gray
et al.
1988).
We assume that there
is
an
upper
limit to the
number of
eagles
that can
be
sup-
ported by
any
stretch of
undeveloped
shoreline.
Thus,
as shoreline continues
to be
modified,
we
believe
that
the
length
of
remaining
undevel-
Table 6.
Distances
(m)
at which
eagles
flushed
in
response
to
approaching
boats
by eagle age,
season,
and
geographic
area,
northern
Chesapeake
Bay, Maryland,
1987-89.
Flush distance
(m)
n
SE
Boat
Ada
40 203.7
16.1
Immaturea
36
228.6 19.1
Summerb
42 175.5
12.1
Winterb
34
264.9
20.6
Eastern Shorec
10 163.0
36.8
Aberdeen
Proving
Grounde
66
223.4
13.0
All
boat
76 215.5
12.4
a
Flush distances
of
adult
and immature
eagles
did not differ
(Wil-
coxon rank-sum
test,
P
=
0.385).
b
Eagle
flush distances
in summer
and winter
differed
(Wilcoxon
rank-
sum
test,
P
=
0.001).
c Eagle
flush distances
on the Eastern Shore
and
Aberdeen
Proving
Ground
differed
(Wilcoxon
rank-sum
test,
P
=
0.049).
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
290
BALD EAGLE-HUMAN
INTERACTIONS
*
Buehler et al.
J.
Wildl.
Manage.
55(2):1991
oped
shoreline
may
become
the
limiting
factor
for
some
eagle
populations,
including
the
Ches-
apeake population.
It
is
possible
that,
as
eagle
populations
con-
tinue
to
grow,
birds
will
adapt
to human
pres-
ence and will
begin
to use
shoreline that is now
avoided.
However,
our data
provide
no
evi-
dence that such
adaptation
is
underway.
There-
fore,
we
believe
that
when
shoreline
is
devel-
oped,
it
is
irretrievably
lost
as
eagle
habitat. In
our
opinion,
the
key challenge
to
eagle
man-
agers
in the
coming
decade,
particularly
in the
eastern United
States,
will
be to ensure the
sus-
tained existence
of
enough undeveloped
habitat
to
provide
for
viable
eagle populations
over the
long
run.
LITERATURE CITED
ANDERSON,
J.
R.,
E. E.
HARDY,
J.
T.
ROACH,
AND R.
E. WITMER. 1976. A
land use and land cover
classification
system
for use with
remote sensor
data.
U.S. Geol.
Surv.
Prof.
Pap.
964.
28pp.
BUEHLER,
D.
A.,
T.
J.
MERSMANN,
J.
D.
FRASER,
AND
J.
K. D.
SEEGAR. 1991a.
Nonbreeding
bald ea-
gle
communal and
solitary roosting
behavior and
roost
habitat
on the
northern
Chesapeake
Bay. J.
Wildl.
Manage.
55:273-281.
~
,
-
,
AND
-
1991b.
Dif-
ferences in distribution of
breeding,
nonbreed-
ing,
and
migrant
bald
eagles
on the northern
Chesapeake Bay.
Condor
93:In
Press.
FRASER,
J.
D.,
L. D.
FRENZEL,
AND
J.
E.
MATHISEN.
1985. The
impact
of
human
activities
on
breed-
ing
bald
eagles
in
north-central
Minnesota.
J.
Wildl.
Manage.
49:585-592.
GRAY,
R.
J., J.
C.
BREEDEN,
J.
B.
EDWARDS,
M. P.
ERKILETIAN,
J.
P. BLASE
COOKE,
O. J.
LIGHTHI-
ZER,
M.
J.
FORRESTER,
JR.,
I.
HAND,
J.
D.
HIMES,
A. R.
MCNEAL,
C. S.
SPOONER,
AND
W.
T.
MURPHY,
JR.
1988.
Population growth
and
de-
velopment
in the
Chesapeake Bay
watershed
to
the
year
2020. U.S.
Environ.
Prot.
Agency,
Ches-
apeake Bay
Liaison
Off.,
Annapolis,
Md.
73pp.
KNIGHT,
R.
L.,
AND S. K. KNIGHT. 1984.
Responses
of
wintering
bald
eagles
to
boating activity.
J.
Wildl.
Manage.
48:999-1004.
MCGARIGAL, K.,
R.
G.
ANTHONY,
AND F. B.
ISAACS.
1991. Interactions of humans and bald
eagles
on the
Columbia
River
estuary.
Wildl.
Monogr.
115:1-47.
SAS
INSTITUTE,
INC. 1986.
SUGI
supplemental
user's
guide.
Version 5 ed. SAS
Inst., Inc.,
Cary,
N.C.
662pp.
STALMASTER,
M.
V.,
AND
J.
R. NEWMAN. 1978. Be-
havioral
responses
of
wintering
bald
eagles
to
human
activity.
J.
Wildl.
Manage.
42:506-513.
Received 12
April
1990.
Accepted
8
October
1990.
Associate
Editor:
Morrison.
A
LOGISTIC MODEL
FOR
THE
CUMULATIVE EFFECTS OF
HUMAN
INTERVENTION ON
BALD
EAGLE
HABITAT
GEORGE J.
MONTOPOLI,
Department
of
Statistics,
University
of
Wyoming,
Laramie,
WY
82071
DONALD
A.
ANDERSON,
Department
of
Statistics,
University
of
Wyoming,
Laramie,
WY
82071
Abstract: We
developed
a
logistic
model based
on
a
conjoint
analysis
approach
to
evaluate the cumulative
effects of selected forms of
human
disturbance
on
bald
eagles
(Haliaeetus
leucocephalus)
in their natural
habitat. The
dependent
variables
were amount of habitat available to bald
eagles
for
foraging
and
perching
as a function of 5
human disturbance factors.
Application
of the
model
to a section of the
Snake
River in
Grand Teton National
Park where detailed information on
levels of human intervention
were
known
yielded
reasonable values
when
compared
to
field observations.
The model
is
currently
in use
by
managers
of the
Bridger-Teton
National
Forest and Grand
Teton
National Park.
J.
WILDL.
MANAGE.
55(2):290-293
In
June
1986,
the
Greater
Yellowstone Eco-
system
Bald
Eagle
Working
Team
(Eagle
Team)
requested
assistance
in
constructing
a cumula-
tive effects model
to evaluate
quantitatively
the
cumulative effects of
human disturbance on
bald
eagles
in their natural habitat.
Development
of
the model
progressed
through
3
stages:
(1)
de-
velopment
of
a
cumulative
effects model
using
a
conjoint
analysis modeling
approach
with a
logistic
regression analysis,
(2)
development
of
a Fortran
computer program
with user's
manual
incorporating
several
managerial analysis
op-
tions
for
implementing
the
model,
and
(3)
val-
idation of the model
with
primary
and
second-
ary
data collected
in
Grand
Teton
National
Park
(NP).
Our
purpose
is
to describe
the basic
modeling
approach,
to
display
the resultant cumulative
effects
model,
and to
summarize results of the
validation
study
conducted
in
Grand
Teton
NP.
This content downloaded from 160.36.178.25 on Tue, 05 Jan 2016 16:17:48 UTC
All use subject to JSTOR Terms and Conditions
