Appendix E. Using the Python scripts in ArcGIS.

After adding the *Hunting Mapper* toolbox (See Supplement 1), double-click the *Map Primate Depletion* script to implement the numerical model (The analytical model can be implemented and analyzed more completely in Matlab (S4), but the script *Map Steady State Primate Depletion* is provided to facilitate map making and the implementation of the model in ArcGis). A window pops up requesting inputs. (1) A raster image of the landscape with units in Meters must be input, and a location and name for the depletion raster output must be specified. (2) Appropriate parameter values must be specified. (3) A table, most easily from a spreadsheet like Microsoft Excel, must be input with the appropriate formatting

FIG. E1. ArcGIS window for inputting parameters.

Each row in the table corresponds to a distinct settlement. At a minimum, the table needs fields labeled āxā and āyā with the UTM x-y values, and fields for the population size in each year to be considered. āpop1ā corresponds to the hunter population (males of hunter age) in year one, and in general āpopnā is the population in year n (Note that when implementing the analytical model, only a single population size should be input for each settlement and the header of the third column should read āpopā since there is no associated year in the analytical model). The specified parameter for the number of years to run the model must be less than or equal to the number of population fields or the script will return an error (Again, with the analytical model there is no year specification because a steady-state solution is generated).

In the table above, the first ten (of fifty) years of settlement history are visible. Originally, there was only a single settlement and over time new settlements appeared. By adding settlements through time the script incorporates settlement fissioning and spread. Projected population growth and spread can also be input based on the population growth model of the users choice. The name field in the table is optional.

FIG. E2. Necessary format of the settlement location and population data.

Note that the diffusion parameter (diffusivity) does not have a field to input. The diffusivity can be changed by opening the script and changing the āDā parameter. The hardwired diffusivity is intended to prevent users from observing traveling waves, which occur due to discretization if the diffusivity is substantially higher than about 0.1. It is possible to allow greater diffusivity without the appearance of traveling waves, but this requires a narrower time window, making the model more computationally expensive.