USING GIS APPLICATIONS TO ESTIMATE WILD TURKEY HOME RANGES AND

1
USING GIS APPLICATIONS TO ESTIMATE WILD TURKEY
HOME RANGES AND HABITAT USE IN NORTHEASTERN SOUTH
DAKOTA
Roger D. Shields, Lester D. Flake, and Chad J.
Kopplin
Wildlife and fisheries Sciences, South Dakota
State University
HOME RANGE ESTIMATION Relocations estimated
by triangulation of 2 4 azimuths from GPSd
telemetry stations. Converted to UTM X and Y
coordinates with Locate, available in CALHOME
(Kie et al. 1996). Relocations imported into
ARCView and used to calculate 90 fixed kernel
home ranges using Animal Movement Analysis (Hooge
and Eichenlaub 1997) (Fig. 5). Within sexes,
differences tested with ANOVA, mixed models, and
non-parametric tests.
HABITAT SELECTION 6 habitats selected for
study (Table 1). Habitats digitized into
vector-based GIS using ARCInfo from aerial
section photos (Fig. 4). Habitat use and
availability measured at 2 scales (Fig. 5).
Figure 6. Habitat selection and avoidance at 2
scales of measurement. Successful nesting
females (top) represent the pattern observed by
groups in general. Error bars ? 1 SE.
INTRODUCTION Knowledge of animal movements and
habitat use is important to management of
wildlife populations. Application of GIS
technology to the analysis of movement and
habitat use data has increased in recent years.
We applied this technology in our study of
eastern wild turkeys (Meleagris gallopavo
silvestris) released in northeastern South
Dakota. Eastern wild turkeys have historically
been considered a forest dwelling subspecies,
requiring large tracts of forest. Recent
introductions have shown the subspecies can
thrive in mixed forest-agriculture landscapes.
The South Dakota Department of Game, Fish and
Parks recently introduced eastern wild turkeys
into Grant County, in northeastern South Dakota.
Grant County offers only minimal woodland habitat
in a highly agricultural landscape. We sought to
determine what effect limited amounts of woody
cover would have on turkey movements and habitat
use.
RESULTS Our estimates 2 10 times larger
than reported for other northern populations of
turkeys. Most similar to home ranges of
populations from forest habitats with poorly
interspersed resources. Home ranges differed
by season for both sexes (males ?24.73, 2 df,
P0.094 females F47.07, 2, 201 df, Plt0.001)
(Fig.?3). Successful nesting females had
larger home ranges than unsuccessful nesting
females (t1.72, 83 df, P0.089).
Home range scale Use habitats within
individual home range Available habitat within
composite availability polygon (created by
merging all individual home ranges) Within home
range scale Use habitats within each buffered
relocation Available habitats within
individual home range
Figure 5. Scales of habitat analysis.
Figure 3. Differences in size of seasonal home
ranges of male and female eastern wild turkeys.
Within sexes, seasons marked with same letter did
not differ (P?0.10). Error bars ? 1 SE.
METHODS 6 males and 67 females fitted with
radio-transmitters were tracked with truck
mounted antenna systems ?3 times/week from Jan
Aug, 1999 and 2000 (Figs. 1 and 2).
SUMMARY Seasonal home ranges were much larger
than reported for populations of wild turkeys in
other northern states, suggesting that resources
are poorly distributed in this area. Home
range size differed by season spring home ranges
were larger than winter and summer ranges, likely
due to increased movement related to reproductive
behavior. Home range size differed by
reproductive class and was associated with
reproductive success. Wild turkeys most often
established home ranges with a woody
cover-pasture complex and avoided the heavily
cultivated landscapes within the study area.
Within home ranges, wild turkey activities were
closely tied to woody cover. Cropland, pasture,
and hayland habitats were used less than
available. Idle grass habitats and farmsteads
and residential areas were used proportional to
availability within home ranges. Brooding
females established home ranges with greater
amounts of cropland, avoiding pasture habitats at
both scales of measurement. During
pre-incubation localization, females successful
in the nesting attempt avoided pasture at both
scales.
Season Designations Designated 3 seasons
(winter, spring, summer) with further
classification of females based on reproductive
status. Within the spring season, a special
pre-incubation, localizing period of initial
nests was studied separately for females.
Thus, 11 groups of turkeys studied (Fig. 3).
Brooding females had smaller home
ranges than non-brooding females (t2.95, 53 df,
P0.005). No difference in home range size
during localizing period (t0.24, 35 df,
P0.815).
Figure 4. Study area digitized into ARCInfo and
ARCView.
Figure 1. Fitting female turkey with
radio-transmitter.
RESULTS Woody cover was highly selected by
all seasonal groups of turkeys at both scales
(Fig. 6). Cropland was avoided at both scales
by all groups, but was utilized in greater
amounts by brooding females. Idle grass and
farmstead/residences were generally used in
proportion to availability, although idle grass
was utilized more by successful localizing
females and brooding females. Pasture was
often selected at the home range scale but
avoided at the within home range scale
successful localizing and brooding females always
avoided pasture habitats. Hayland was
generally used in proportion to availability at
the home range scale while it was proportional or
avoided at the within home range scale.
Figure 2. Truck mounted antenna system.
AKNOWLEDGMENTS