Edwards and Eberhardt (1967) reported the results of a live-trapping study on a penned population of 135 cottontail rabbits. The wild cottontails were captured and placed in a 40-acre rabbit-proof enclosure. After allowing 4 days for the rabbits to adjust to their new surroundings, livetrapping was conducted for 18 consecutive days. The raw data are located in a file named rabbit.cap. We are going to use program 2CAPTURE to create an input file named captur.inp. We will use this input file and program CAPTURE to estimate population size, calculate confidence intervals, and perform tests of closure and equal-catchability assumptions.
Use a text editor to view the raw data file named rabbit.cap. Note the format of the data (columns, spacing, binary code) - this is known as X matrix format.
View the file peromysc.cap; these data are in X-Y reduced format (i.e., the data include information on x-y coordinates (trap location) of each capture for each animal).
View the file microtus.inp . These data also are in X-Y reduced format; however, notice the "input" file contains command lines that will be operated on by program CAPTURE.
We will use program 2CAPTURE to convert rabbit.cap to an input file that contains command lines..
Start program 2CAPTURE by typing 2capture at the DOS prompt (make sure you are in the right directory!). The main menu will appear, with a button-bar at the top of the screen.
Select Title from the button-bar menu and enter "Rabbit Data" when prompted for a title.
Select Read . Another menu list will appear.
Select Occasions. This refers to the number of trapping occasions. Note: press F1 for information on a menu item. How many trapping occasions were there in this example?
Because data for this example do not include information on capture by trap coordinate (X-Y data), we cannot select Summary statistics.
Select Input Type . Select X matrix from the list.
The program will ask if you want to read the matrix from a file? Type Y.
The program will then ask you for the name of the input file. Type rabbit.cap .
It will ask if you would like to see the input data? Type Y for yes (notice the column number and range of the data). Exit the view screen by typing "x" (see prompt at bottom of screen).
The program will then ask for the file format. This refers to the column numbers and type of data in those columns (see item 9 above). Enter the following: (T15, A3, 3X, 18F1.0). This tells the program: (1) skip to column 15, (2) there is an alpha-numeric identification in columns 15-17, (3) the next 3 columns are blank, (4) capture data consist of 1 character or number, begin in column 21, and continue for 18 columns (corresponding to trap day). Note: parentheses around the file format are required!
Return to the main menu (use the escape key).
Select Testing.
Select Model Selection (notice that it inserts a "task command" into the input file)..
Do not select Occasions at this time. Note: this option is used if you want to restrict the analysis to specific trapping occasions, e.g., trap days 1-5 (versus 1-18).
Select Closure Test. This option requests a test of the closure assumption. Type N for no occasions.
Do not select Density. Note: this option is only available if trapping data include x-y coordinates (i.e., location of traps in a grid).
Return to the main menu and select Estimation. This tells program CAPTURE what analysis options to use.
Select Abundance.
Do not select Occasions (not applicable in this case - we are using all trapping occasions)
Select Estimators.
Select Appropriate (this option tells program CAPTURE to select and use the most appropriate model to estimate abundance). Note: you may select >1 estimator. However, to keep things simple, choose "appropriate" to let the program select the model to use for estimation.
Return to the main screen (press Esc until you are back at the main screen).
If everything is okay, select Execute program under Execute.
NOTE: if your computer does not have enough working memory, program CAPTURE cannot be successfully invoked from within 2CAPTURE. You will get an error message to that effect (unable to spawn process press any key). Nevertheless, selecting "execute" will cause 2CAPTURE to create the input file captur#.inp.
If you experience the problem just described, exit 2CAPTURE and run CAPTURE in batch mode:
Use the Start tab to enter Programs and MS-DOS prompt
Move to the G: drive by typing cd.., enter, G:, and enter
type at the DOS prompt G:\Capture> capture i=captur1.inp o=rabbit1.out
type exit to leave DOS mode
Remember, o=rabbit1.out tells the program to send the output to a file.
If you were successful in running CAPTURE from within 2CAPTURE, output will be sent to a file named captur#.out.
If you are running CAPTURE in batch mode and the analysis was executed successfully, the following statement will appear on your screen: Execution Terminated: Capture normal end. Unsuccessful runs read Execution Terminated: Capture Abort. |
Use any text editor to view the output (e.g., type edit rabbit1.out at the DOS prompt).
View microtus.inp (a batch file, complete with data and command lines). Note the following:
data are in xy-reduced format (i.e., captures were recorded for each animal by trap location and trap day),
SUMMARY option has been requested, which provides information on distance between captures, recaptures, etc. (Note: you can only request the summary option if you have XY data).
a different format statement is used (A3,36(F2.0)) -compare it to the rabbit input file. Remember that the microtus and peromyscus data include XY coordinates.
Uniform Density Test option has been requested. This produces a grid of captures as a first step in evaluating the uniform-density hypothesis (important for density estimates). You will not need this information to answer questions from the problem set, but it is interesting to see how the captures were distributed over space.
analysis is restricted to trapping occasions=4-8. Note: For initial analyses, like your problem set, use all the trapping occasions.
Run CAPTURE in batch mode using the input file named microtus.inp, i.e., at the DOS prompt type: capture i=microtus.inp o=microtus.out .
View the output. Take a look at the additional information provided by the Summary and Density options.
We will discuss interpretation during class.
Revised: 25 August 2011