Bryan
Hummel
Ecology 3434
Ribble;
8:30 TR
March
16, 1999.
Introduction:
A question was raised about the
reaction of small mammal populations in areas where humans have changed the
habitat of Edwards Plateau woodlands by various clearing techniques. A study was developed to estimate the
population of these mammals in the various treated areas and then compare these
to the numbers in the untreated areas.
Methods:
An area of Emilie and Albert
Friedrich Park was selected near the Vista Loop Trail because of the close
proximity of three different types of habitat.
There were untreated areas composed of Ashe Juniper and Oak woodlands next
to areas treated by the clearing of the Junipers without fire and also next to
areas that were cut then treated with fire.
There were ten transects divided between the three habitats. Ten live-catch Sherman traps were set up on
a ten-meter interval for each transect.
Transects C and F were set up in the untreated Juniper and Oak woodlands
and used as the control. Transects D,
G, I, and J were set up in the cut unburned areas while A, B, E, and H were set
up in the cut areas treated with fire.
The Lincoln – Peterson mark and recapture method was used estimate the
population for each area. The results
were then compared to find which types of habitat were preferred by the areas
most commonly caught small mammal, Peromyscus
pectoralis.
Results:
The populations of each of the ten
transects were calculated individually then they were grouped together by
habitat. The results of the transects
combined by habitat are more conclusive, because several of the individual
transects had insufficient captures to get good population estimates.
|
Transect |
Origional
(M) |
Total # in
2nd (n) |
Recapture
(m) |
Total
Population (N) |
Std. Error |
(+/-) 95%
Confidence |
||
|
A |
1 |
1 |
1 |
1 |
0 |
0 |
||
|
B |
0 |
0 |
0 |
#DIV/0! ( 0 ) |
#DIV/0! |
#DIV/0! |
||
|
C |
1 |
1 |
1 |
1 |
0 |
0 |
||
|
D* |
0 |
2 |
0 |
#DIV/0! ( 2 ) |
#DIV/0! |
#DIV/0! |
||
|
E |
2 |
1 |
1 |
2 |
0 |
0 |
||
|
F |
1 |
1 |
1 |
1 |
0 |
0 |
||
|
G |
2 |
2 |
1 |
4 |
8 |
15.68 |
||
|
H |
0 |
0 |
0 |
#DIV/0! ( 0 ) |
#DIV/0! |
#DIV/0! |
||
|
I* |
2 |
2 |
0 |
#DIV/0! ( 4 ) |
#DIV/0! |
#DIV/0! |
||
|
J |
4 |
4 |
2 |
8 |
16 |
31.36 |
||
|
|
|
|
|
|
|
|
||
|
Totals |
13 |
14 |
7 |
26 |
48.28571 |
94.64 |
||
|
"Native" |
2 |
2 |
2 |
2 |
0 |
0 |
||
|
Cut * |
8 |
10 |
3 |
26.66666667 |
165.9259 |
325.2148148 |
||
|
Burned |
3 |
2 |
2 |
3 |
0 |
0 |
||
|
|
|
|
|
|
|
|
||
|
Cut
Habitat with P. pectoralis only |
8 |
8 |
3 |
21.33333333 |
94.81481 |
185.837037 |
||
Table 1. March 1999 trapping results from Friedrich
Wilderness Park, Bexar County, Texas.
The results are by transect and also by habitat for the original number
of rodents marked and captured the first day, the total number caught the
second day and the number of rodents recaptured the second day. The total population was calculated using
the Lincoln – Peterson method. The
standard error and 95% confidence interval were calculated. The “#DIV/0!”
symbol is shows where the model proposed dividing by zero which is
mathematically impossible and thus no conclusive data can be gathered in these
cases. Transects D and I both contain
species other than Peromyscus pectoralis
in the totals for the second day, (Peromyscus
atwatteri and Sigmodon hispidus
respectively). The last row takes this
into account and provides the information on Peromyscus pectoralis only.
The
information collated in Table 1 shows that because of the large standard errors
and 95%confidence intervals, the data cannot be considered statistically
different for any of the three areas.
Discussion:
The assumptions of the Lincoln – Peterson
index were satisfied at Freidrich Wilderness Park even though the data obtained
was not conclusive. The individuals
were randomly sampled and because of the short (24 hour) time frame there was
little if any migration or mortality, and little time for the rodent to lose
its tag. To decrease the possibility
for error, several things can be done such as setting more traps in each
transect or adding more transects in similar habitats. With larger numbers of mammals captured and
most likely more recaptures, the possibility for errors decrease and so too may
the standard error. A larger sample
size may allow for better analysis of the data.
Bryan Hummel
Ecology 3434
Ribble; 8:30 TR
March 16, 1999.