WebSim provides a windows interface for all of the necessary input to run a simulation. The command line interface provides a few extensions to the features of the windows interface, but is not needed for the vast majority of applications. When the WebSim interface is opened, the input window opens automatically. The input window is controlled through a tab strip at the top of the window.

Habitat Tab

Under the habitat tab, the user must enter a scale parameter. The scale is a multiplier carried through the entire simulation that allows the user to adjust the size of the area on which the trapping web and animals are placed. A scale of 1 creates an area of 251,001 square units per web.

Three file names can be set at the habitat tab. First, the animal location file provides the initial location of all the animals in one web. Second, the trap location file provides the locations of the traps. Finally, the results file contains all of the simulation results. All three of the files will default to the names shown if no changes are made.

Animals Tab

First, under the animals tab is the population setting. The population size entered is the average number of animals in the area of each web. Then interface will display the density which corresponds to the value entered. The actual number of animals per web is a Poisson random variable with a mean of the number entered.

Model tbh may be selected by clicked on the Mtbh check box. When the check box is selected the frame containing the tbh parameters becomes active. In the tbh frame, enter an average capture probability, and the a and b parameters for the beta distribution from which individual capture probabilities are derived. Currently, only integer values for a and b may be entered in the windows interface, but any value may be entered at the command line. See the detailed description of the algorithm used to calculate capture probabilities.

To select an animal distribution click on its corresponding radial. Three types of animal distributions are available. Under a uniform random distribution each animal is assigned a location from a uniform distriution. For a clustered random distribution, the location of a cluster is randomly selected from a uniform distribution and the cluster size is pulled from a Poisson distribution. If a clustered distribution is selected, enter a mean cluster size which becomes the l parameter of the Poisson distribution.With a uniform home range distribution, animals are evenly placed across the habitat.

Next, select a movement type. Under random walk movement regime, a distance and direction are selected from a probability distribution and the animal moves to the new location without bounds of a home range. Within home range movement only allows animals to move within a defined radius of their initial location. The probability that an animal moves at any given time step is determined by probability of movement parameter. Two movement distance distributions are available. The distance may be pulled from an exponential distribution, in which case the mean distance is entered, or the distances may be pulled from a uniform( 0, maximum distance) distribution, in which case the maximum distance is entered.

Traps Tab

Under the traps tab the user defines the size and design of the trapping web. The web may be input in two ways: 1) enter the number of lines, number of traps per line, and the number of traps at the center. The interface calculates the total number of traps for you. 2) enter the number of lines, number of traps at the center, and the total number of traps. The interface then calculates the number of traps per line.

The distance between traps along a line must be entered at the trap spacing box. Once entered, the interface will calculate the length of a line and the area covered by a web.

Finally, select whether the traps can catch a single animal (i.e. a Sherman trap or snap trap) or can catch multiple animals (i.e. a pit fall trap).

Simulation Tab

Specifics about the simulation are entered under the simulation tab. First, enter a random number seed, 0 seeds from the system clock. Second, enter the number of replicates. Next, enter the number of trapping occasions. Then, enter the number of time steps per occasion. The time steps are a way to approximate continuous time with a discrete algorithm. Remember animals move up to a maximum of the distance entered under the animal tab at each time step. Next, enter the number of webs. A maximum of 50 webs may be set per simulation. The number of animals near each web varies as a Poisson random variable. Then, enter the catch distance. The catch distance is the radius around a trap within which an animal has a non-zero probability of capture. The remaining options all have defaults to common and well justified values. The user may change any of them if desired.

Five options are available for density estimation using distance methods. Three models may be run individually: 1) hazard-rate with simple polynomials, 2) half-normal with Hermite polynomials, and 3) uniform with cosine adjustment (Fourier series). In addition, two sets of model averaged estimates are available. The first set combines the hazard-rate, half-normal, and uniform models. The second combines the half-normal and uniform models. Beware when selecting the hazard-rate model because it is sensitive to "spuriously spiked data", where there are more observations near zero distance than expected. The hazard-rate may produce erroneous estimates in such situations, despite fitting the data very well.

For more information on estimation, model selection, or variance components, refer to the Distance Sampling online book.

Density may also be estimated with the Link and Barker (1994) geometric estimator. Click the check box for the Link and Barker estimator and input fields appear. The comment field must not contain any spaces. K is the number of parameters. The starting values are for the optimization routine.

Click Run Simulation to begin the simulation.