|
|
Evaluation of a localized white-tailed deer management
technique with the use of capture data
Jason Isabelle
INTRODUCTION
The Huntington Wildlife Forest (HWF) is a 6000-ha research forest located
5 km west of Newcomb, New York, on which a variety of studies have been
conducted; the goal of many being to increase our knowledge of the biology,
behavior and movements of white-tailed deer (Odocoileus virginianus) .jpg)
.
With white-tailed deer populations increasing rapidly in urban/suburban
environments all throughout the country, it is evident that we must find
a way to keep their numbers at balanced levels. Many of the studies conducted
at HWF were done so in an attempt to discover such a management technique.
In one such study, a single social group comprised of 14 deer was removed
from the Southern end of HWF and relocated to another part of New York
State. After the removal, which occurred in 1994, the major questions
that faced researchers were; will deer from surrounding areas move into
the area that the deer were removed from, (also known as the "void"
)
or will the area of removal remain vacant for a number of years, and if
so, for how many? If the void stayed unoccupied for a long period of time
after the removal, researchers hypothesized that this type of management
would be an effective way to control deer populations on a small scale.
Some people might think that deer from surrounding areas would move into
the void because there would be more resources available to them. Others
might think that the area would remain unoccupied because the deer from
surrounding areas would be preferential to the land that they were born
on and would not want to move into an unfamiliar area. The focus of my
research was to answer the questions listed above.
BACKGROUND
In 1985 a study conducted at HWF showed that female fawns do not disperse
away from their mother's home range, but instead set up their home ranges
adjacent to and often overlapping that of their mother's. 
Because
of this strong tendency of female fawns to live near their mothers, the
home ranges of a population of deer are not randomly located on the landscape.
If we were to picture a piece of white-tailed deer habitat and more specifically
the home ranges of the deer that appeared on the landscape, we would see
that they were organized in a very distinct and structured manner. The
home range of the oldest doe that started the population would be located
in the center of the landscape and her female offspring's home ranges
would boarder/overlap hers. As these deer gave birth to more female fawns,
the home ranges of these deer would be set up beside that of their mother's.
The term "social group" is the term that is given to the group
of deer that was described above. A social group is defined as the deer
that occupy the same geographical range and have similar seasonal movements.
The rose-petal metaphor was used to describe this arrangement of deer
because, as a population of deer expands, the overlapping home ranges
of the female deer look like the petals of a rose 
.
Bucks were not important in this study due to the fact that they disperse
rather than set up home ranges close to their place of birth. Studies
have shown that 100% of all bucks will disperse by the age of 2.5, whereas
this percent is much smaller in female deer (3%).
HWF contained 8 well-studied social groups before one such group (the
target group), consisting of 14 individuals, was removed in 1994. This
was done in order to monitor the response of other surrounding social
groups to their abs
ence.
The target group actually consisted of 17 individuals but researchers
were unable to capture 3 of these deer. Researchers hypothesized that
the removal of a social group from a landscape, or all deer in a relatively
small geographic area (< 2 square kilometers) would create an area
with low deer densities. They also hypothesized (with the use of computer
models) that this area of low density could persist for as long as 10
years due to the tendencies of female deer to live near the area where
they were born .
The researchers, using modified Stephenson box traps (baited with salt),
roadside observation data, and radiotelemetry, monitored the void for
2 years after the social group was removed in order to test this hypothesis
. It was confirmed that a low density area was 
indeed
created when the 14 deer of the target group were removed. It was discovered
after 2 years of observation that no deer located outside of the low density
area shifted their home ranges significantly closer to the void. Following
this research, white-tailed deer on HWF continued to be trapped with the
use of Stephenson box traps, and monitored with the use of radiotelemetry.
This trapping and movement monitoring has continued on a yearly basis.
METHODS
My independent research involved looking at the capture data from 1990-2000
in an attempt to see how much of an effect the removal of the 14 deer
had on the deer densities inside of the void. I was also trying to determine
if the capture rates were significantly different between post-removal
years inside of the void, which would indicate whether or not deer were
repopulating this area. I began my research by examining the capture data
from the years of interest (1990-2000). I tallied the number of captures/trap
by year. When tallying the number of captures, I counted only female deer
(not bucks) and I only counted each captured deer once/year. For example,
if an individual deer was caught 5 times in one year, only the first capture
of this deer was taken into account. Through all of my calculations, I
always kept the data collected from traps inside the void separate from
the data that was collected outside the void. Data from different years
was also kept separate unless a sum value was needed for pre (1990-1994)
or post-removal (1995-2000), in which case the years were lumped together
accordingly. After examining the capture data from the years listed above,
I then looked at the trap record data, which told me the number of nights
that each trap was set. By now dividing these two numbers (captures and
nights set), I was able to figure out which traps caught the most deer
(the units for this calculation were; the number of deer caught/night
that the trap was set). To make this a larger number that was more meaningful
and easier to work
with, I multiplied it by a factor of 100 (the units for this calculation
were; the number of deer caught per 100 nights that the trap was set).
When finished with my calculations, I had come up with the average number
of deer caught per 100 trap nights for every trap for every year between
1990-2000. I then averaged the capture rates for traps located inside
of the void (each year separately); this was also done for traps located
outside of the void (each year separately). These numbers were then averaged
for the years before the target group was removed (1990-1994); this was
also done for the years after the deer were removed (1995-2000).
After all of these calculations were completed, I used a General Linear
Model (GLM) in the SASä computer program to manipulate the data that
I had just collected. I ran 8 different trials with SASä, which helped
me answer questions that I had pertaining to my research. These trials
helped me compare data between several factors. The factors that I was
most interested in comparing were, the void and the area outside of the
void as well as between pre and post-removal years.
RESULTS
Although I found a significant decrease in the capture rates between the
pre and post removal years inside of the removal area, I also found a
significant decrease outside of the removal area as well. I also found
that there was a significant difference between capture rates in the post
removal years inside of the void (
comparing
year to year), but this was not the case outside of the void. When the
post removal years were compared between the void and the non-void areas
though, the difference was not significant. When I compared the capture
rates inside of the void in pre-removal years with those outside of the
void during the same time period, I found them to be insignificant. When
I looked at each year separately (1990-2000) between the void and the
non-void, 1993 was the only year that I found to be significantly different
.
DISCUSSION
Although most of my crucial statistical analyses turned out to be insignificant,
I concluded that there were too many outside factors affecting the research
to make a definite conclusion of whether or not we were accurately monitoring
what was going on with the deer population on the Southern end of HWF.
One major factor that made interpreting the results difficult was the
general decreasing trend in the deer population on HWF. By examining the
attached graph and data table, I concluded that the deer population on
HWF had decreased between 1990-2000. This drop in deer numbers was probably
due largely to the heavy amounts of winter mortality that occur in the
Adirondacks. When snow levels reach roughly 38cm, deer at HWF leave the
property and travel to their winter range where snow depths are less and
they can get out of the cold winter winds. The number of days that deer
spend on their winter range is often used to judge the severity of the
winter. I averaged the number of days that deer spent on winter range
for pre-and post-removal years and found that this number in post-removal
years is almost double what it was in pre-removal years. If the severity
of the winters had not increased so dramatically in post removal years,
the non-void deer densities would have probably stayed the same while
the densities in the void would have decreased. The removal of the 14
deer from the void might very well have lowered deer density, but because
of the decrease in the population in general it was difficult to see exactly
how the removal affected the deer densities.
Several of the other factors that contributed to the difficulty of interpreting
my results were associated with the actual removal of the deer from the
property. For example, there may have been a few deer of the target group
that spent some time outside of the void as well as inside of the void,
therefore it would be understandable that capture rates would drop in
both places in response to their removal. There are also several factors
associated with using capture rates as an index of population size that
made it difficult for me to make an accurate conclusion about the findings
of my research. First of all, some deer are more cautious than others
and therefore are less susceptible to being trapped. New traps were also
built and replaced all throughout the 10 years that I looked at; some
located in the void and some out of the void. During certain years there
were more new traps outside of the void than inside the void and vice
versa, thus affecting the capture rates. Deer seem more reluctant to go
near new traps probably due to the fact that the traps have a strong odor
of new wood and water sealant. With all of these factors combined, it
brings up the question of whether or not trapping is a sensitive enough
method to warrant its use as an index, especially at a place like HWF
where deer densities are relatively low to begin with.
Despite the insignificance of many of my statistical tests, there were
some trends in the data that suggest that the removal did cause a decrease
in the capture rates. If you look at the graph that I have attached, you
can see that from 1993-1998, the capture rates outside of the void stay
in the same general range, while the capture rates from the void drop
substantially. If you also examine the graph and look at 1995 (the year
after the deer were removed) you can see that the capture rate in the
void was considerably lower than it was in the area located outside of
the void. There is also a substantial amount of data that supports the
fundamental nature of the deer biology that would make this method of
management a success, these being; the philopatry ("love of homeland")
that female white-tailed deer have, coupled with their low dispersal rates.
Support can also be seen in genetics studies, as well as those concerned
with deer behavior. These studies suggest that interactions, between populations
of deer in close proximity, are limited in some environments. Studies
have also been conducted in other areas of the country where northern-forested
regions prevail; deer were shown to possess a great deal of philopatry
here as well.
MANAGEMENT IMPLICATIONS
Although my analysis of the data was inconclusive, this by no means rules
out the possibility of using this form of localized management as an effective
way of reducing deer populations to desired levels. Although the white-tailed
deer is a magnificent animal that is enjoyed by many, we cannot ignore
the great amounts of damage that they create in urban/suburban areas.
Deer in these areas cause millions of dollars in damage to plants and
shrubs alone. Deer also pose a risk to human health in a couple of ways.
First of all, deer/vehicle collisions kill a number of people each year
and injure many more, as well as causing a great deal of money loss to
the owners of the vehicles with which they collide. Deer are also the
carriers of the primary vector of Lyme disease (the deer tick) that has
caused numerous health problems all along the eastern seaboard. With white-tailed
deer populations becoming overabundant in many areas of the country, especially
in urban/suburban areas, a management technique suitable for these types
of environments seems more important now than ever before. With a great
percentage of this countries' deer located in urban areas including parks,
it has been very difficult, if not impossible to implement traditional
management techniques due to their high cost and impracticality in these
areas. Deer cannot usually be hunted in these areas either, due to their
close proximity to humans. Localized management has important implications
for these areas, because it can be implemented on a small scale and does
not involve as many deer as traditional management practices do. It is
also for this reason that localized management is often a more socially
acceptable form of management. Localized management could be used in a
number of different situations, but probably would be implemented most
often by wildlife biologists. With localized management, biologists can
concentrate on removing deer from a small area and creating a low density
area, rather than broadening their efforts to a larger area, which is
often unsuccessful.
.jpg)
ABOUT
THE AUTHOR: Jason Isabelle
A.S. Paul Smiths College (Ecology and Environmental Technology); currently
an undergraduate student (senior) enrolled in Environmental and Forest
Biology at SUNY-ESF. My interests include primarily the management of
game species, in particular, the white-tailed deer and the wild turkey.
After receiving my B.S. in the spring of 2002, I would like to attend
grad school and further pursue my education. Upon completion of my schooling,
I hope to become a state wildlife biologist, where I will work to ensure
that our natural resources will be around for future generations to enjoy.
Contact Information
Jason Isabelle; E-mail jlisabel@syr.edu
|
|