Population Statistics I
Homework 4
This homework assignment consists of several problems. Each problem is a
capture-recapture experiment. Perform the required analyses, then for
each problem, write a separate executive summary of at most 1 page for each
problem. I'm not interested in seeing output generated by the various
programs. I want to see a report from you on what you believe to be the
important and relevant details of your work. Obviously it is important
to report on what software you used or how you computed your results.
Problem 1.
Rose Lake Wildlife Research Station in Michigan conducted yearly surveys
of cottontail rabbit populations prior to the hunting season. Several years
of this data are given in Bishop et al. (1975:245) where the data are presented
as two trapping periods and a single harvest sample. The capture history
data for 1968 are given in the following table.
First
Sample | Second
Sample | Hunter's
Bag | Count |
|
|
|
|
| Present | Present | Present | 9 |
| Present | Absent | Present | 3 |
| Absent | Present | Present | 4 |
| Absent | Absent | Present | 41 |
| Present | Present | Absent | 31 |
| Present | Absent | Absent | 41 |
| Absent | Present | Absent | 29 |
| Absent | Absent | Absent | . |
- Using a log-linear models approach, fit closed population models and estimate
the rabbit population size. Report on the the models that you tried and why you
selected the model that you have reported as best. Give the population estimate
and confidence interval estimate.
- Use the MARK (and CAPTURE if you like) software to analyze the data as a
closed population. Again report on the models that you tried and why you selected
the model that you have reported as best. Give the population estimate
and confidence interval estimate. Take the above table of data and start by creating
an INP file of capture histories with frequencies.
Problem 2.
Cormack (1989) reports on a data set of Eider ducks and gives the capture
history data for 6 periods of time (the first and last times are actually
data pooled over more than 1 year). Treat this data set as having come
from 6 yearly capture-recapture samples. The
EiderDucks.inp file is already
in a format suitable for use with program MARK.
- Fit open population models to these data. Assume that there is no
emigration/immigration during the study. Provide a graphic showing
estimates of survival during the study period (for intervals that can
be estimated). Report on your model selection strategy and why you
selected the model to report on that you did. Give a table of survival
estimates along with standard errors and confidence limits.
BONUS POINTS (10): Use program JOLLY (or spreadsheet or other program)
to compute Jolly-Seber estimates of population
size, recruitment, and survival during the study period. Report these
results on a separate page with estimates, standard errors, and
confidence limits.
Problem 3.
Brownie et al. (1985) report on adult Mallard duck band return data
from San Luis Valley. Fit band recovery data models to these data.
Report on the model that you have selected as a "best" fit, and
tell why you selected that specific model. Give a table of survival
estimates, standard errors, and confidence limits. The data are
given in the table below in the recovery matrix input format
that can be used directly by program MARK. Place this data into
an ".inp" file and use it with MARK to perform the appropriate
analyses.
/* San Luis Valley Adult Mallards: Page 92, Brownie et al. 1985 */
glabel(1)=Adults;
recovery matrix group=1;
10 13 06 01 01 03 01 02 00;
58 21 16 15 13 06 01 01;
54 39 23 18 11 10 06;
44 21 22 09 09 03;
55 39 23 11 12;
66 46 29 18;
101 59 30;
97 22;
21;
231 649 885 550 943 1077 1250 938 312;
To Top of Document
Last Modified: April 10, 2001