SCATMAN - User Instructions

Program "SCATMAN" : User Instructions

James E. Hines and William A. Link

USGS, Biological Resources Division
Patuxent Wildlife Research Center
11510 American Holly Drive
Laurel, Maryland 20708

Introduction

This program is used to test the hypothesis of prey-selectivity based on random samples of predator scats. Suppose that prey species i has population density d_i, and that a single kill of species i typically results in λ_i scats. Under the null hypothesis of no prey specificity (i.e., that the frequencies of prey species in a predators' diet is proportional to availability), the proportion of scats identified to species i is expected to be

     d_iλ_i
π_i= ------
    ∑ d_iλ_i
    ⁱ

If the quantities d_i and λ_i are known, a chi-squared goodness of fit test can be used to test the hypothesis of no prey specificity (Manly et al. 1972, Chesson 1978). Typically, however, d_i and λ_i are estimated rather than precisely known; furthermore, the number of scats produced by a single kill is variable. These sources of variation cause an inflation of the Type 1 error rate of the hypothesis test (Link and Karanth 1994). This program implements a parametric bootstrap designed to handle these problems; for details, see Link and Karanth (1994).

Data input:

The input file for program SCATMAN consists of a list of prey species, a list of estimated densities, standard errors of estimated densities, scat production rates, and scat production variability. The user also specifies the number of bootstrap replications.

Example (Tiger scat data from Link and Karanth, 1994):

Sample data

Prey:	SBR	PIG	CHT	GAR	MJK	LGR
Scat Freq:	118.5	35	143	82.5	24	16
Scat Production:	22.55	11.48	14.09	23.87	7.46	3.54
SD(Scat Production):	9.02	4.59	5.64	9.55	2.98	1.42
Density:	1.80	1.28	5.94	0.84	3.80	4.77
SE(Density):	0.62	0.23	0.81	0.34	1.55	0.63

Explanation:

In this example, the predator species is the tiger (Panthera tigris); prey species are

SBR: Sambar (Cervus unicolor)
PIG: Wild pig (Sus scrofa)
CHT: Chital (Axis axis)
GAR: Gaur (Bos gaurus)
MJK: Muntjac (Muntiacus muntjak)
LGR: Hanuman Langur (Presybtis entellus)

Input file: (scatman2.inp):

200,54321
SBR
118.5,1.8,0.62,22.553
PIG
35,1.28,.23,11.48
CHT
143,5.94,.81,14.085
GAR
82.5,.84,.34,23.867
MJK
24,3.8,1.55,7.463
LGR
16,4.77,.63,3.54

The line containing 200,54321 tells the program how many bootstrap replications (200) to perform, and a seed number (54321) for the random number generator. You probably don't need to change these.

The next line is the first prey species name (SBR). The following line contains the scat frequency for SBR (118.5), the density for SBR (1.8), the standard error of the density (0.62), and the scat production (22.55).

The other pairs of lines contain the data for the other 5 prey species.

Program Operation:

The first step is to create an input file. The easiest way to do this is to look at the sample input file provided, scatman.inp. Use any editor and create an input file similar to the sample input file. Just make sure to save the file as ASCII text. Next, run the program as described below, then view the output using your editor. If you are running Windows, open a MSDOS window and type the following:

cd \scatman
scatman <scatman.inp >scatman.out

Variability in scat production rates was set at 40% of the mean level.

Program Output

bootreps=         200 ix=       54321
 SBR       118.5000  1.8000  0.6200 22.5530
 PIG        35.0000  1.2800  0.2300 11.4800
 CHT       143.0000  5.9400  0.8100 14.0850
 GAR        82.5000  0.8400  0.3400 23.8670
 MJK        24.0000  3.8000  1.5500  7.4630
 LGR        16.0000  4.7700  0.6300  3.5400

                          estimated
species        observed    expected
~~~~~~~       ~~~~~~~~~~  ~~~~~~~~~~
 SBR           118.5000     83.2785
 PIG            35.0000     30.1445
 CHT           143.0000    171.6323
 GAR            82.5000     41.1276
 MJK            24.0000     58.1772
 LGR            16.0000     34.6400

                 CV(SCATRATE)=          10%
***********************************************************
                  Chi     Unadjusted    Adjusted   standard
species         squared      p-value     p-value      error
~~~~~~~        ~~~~~~~~   ~~~~~~~~~~    ~~~~~~~~   ~~~~~~~~
 SBR            18.5917      0.0000      0.0010      0.0003
 PIG             0.8427      0.3586      0.3887      0.0024
 CHT             8.0907      0.0044      0.0539      0.0046
 GAR            46.1486      0.0000      0.0000      0.0000
 MJK            23.3153      0.0000      0.0003      0.0001
 LGR            10.9342      0.0009      0.0027      0.0002

 Composite      92.1822      0.0000      0.0000      0.0000

                 CV(SCATRATE)=          20%
***********************************************************
                  Chi     Unadjusted    Adjusted   standard
species         squared      p-value     p-value      error
~~~~~~~        ~~~~~~~~   ~~~~~~~~~~    ~~~~~~~~   ~~~~~~~~
 SBR            18.5917      0.0000      0.0026      0.0006
 PIG             0.8427      0.3586      0.4043      0.0035
 CHT             8.0907      0.0044      0.0536      0.0054
 GAR            46.1486      0.0000      0.0000      0.0000
 MJK            23.3153      0.0000      0.0009      0.0003
 LGR            10.9342      0.0009      0.0030      0.0002

 Composite      92.1822      0.0000      0.0000      0.0000

                 CV(SCATRATE)=          30%
***********************************************************
                  Chi     Unadjusted    Adjusted   standard
species         squared      p-value     p-value      error
~~~~~~~        ~~~~~~~~   ~~~~~~~~~~    ~~~~~~~~   ~~~~~~~~
 SBR            18.5917      0.0000      0.0042      0.0010
 PIG             0.8427      0.3586      0.4227      0.0052
 CHT             8.0907      0.0044      0.0812      0.0094
 GAR            46.1486      0.0000      0.0000      0.0000
 MJK            23.3153      0.0000      0.0011      0.0003
 LGR            10.9342      0.0009      0.0044      0.0004

 Composite      92.1822      0.0000      0.0000      0.0000

                 CV(SCATRATE)=          40%
***********************************************************
                  Chi     Unadjusted    Adjusted   standard
species         squared      p-value     p-value      error
~~~~~~~        ~~~~~~~~   ~~~~~~~~~~    ~~~~~~~~   ~~~~~~~~
 SBR            18.5917      0.0000      0.0062      0.0018
 PIG             0.8427      0.3586      0.4402      0.0064
 CHT             8.0907      0.0044      0.1268      0.0128
 GAR            46.1486      0.0000      0.0000      0.0000
 MJK            23.3153      0.0000      0.0025      0.0007
 LGR            10.9342      0.0009      0.0056      0.0006

 Composite      92.1822      0.0000      0.0000      0.0000

REFERENCES

Chesson, J. (1989). The effect of alternative prey on the functional response of Notonecta hoffmani. Ecology 70: 1227-1235.

Link, W.A. and Karanth, K.U. (1994). Correcting for overdispersion in tests of prey selectivity. Ecology 75: 2456-2459.

Manly, B.F.J., Miller, P., and Cook, L.M. (1972). Analysis of a selective predation experiment. American Naturalist 106: 719-736.

If you have questions, problems or comments with this program please contact:

Jim Hines, USGS-PWRC
11510 American Holly Dr. #201
Laurel, Md. 20708-4017

Phone:(301)497-5661
email:Jim_Hines%40usgs.gov
web :http://www.mbr-pwrc.usgs.gov/software