Non-Linear Density Dependence in a Stochastic Wild Turkey Harvest Model

Current eastern wild turkey (<I>Meleagris gallopavo silvestris</I>) harvest models assume density-independent population dynamics despite indications that populations are subject to a form of density dependence. I suggest that both density-dependent and independent factors operate simultaneously on wild turkey populations, where the relative strength of each is governed by population density. I attempt to estimate the form of the density dependence relationship in wild turkey population growth using the theta-Ricker model. Density-independent relationships are explored between production and rainfall and temperature correlates for possible inclusion in the harvest model. Density-dependent and independent effects are then combined in the model to compare multiple harvest strategies. To estimate a functional relationship between population growth and density, I fit the theta-Ricker model to harvest index time-series from 11 state wildlife agencies. To model density-independent effects on population growth, I explored the ability of rainfall, temperature, and mast during the nesting and brooding season to predict observed production indices for 7 states. I then built a harvest model incorporating estimates to determine their influence on the mean and variability of the fall and spring harvest. Estimated density-dependent growth rates produced a left-skewed yield curve maximized at ~40% of carrying capacity, with large residuals. Density-independent models of production varied widely and were characterized by high model uncertainty. Results indicate a non-linear density dependence effect strongest at low population densities. High residuals from the model fit indicate that extrinsic factors will overshadow density-dependent factors at most population densities. However, environmental models were weak, requiring more data with higher precision. This indicates that density-independence can be correctly and more easily modeled as random error. The constructed model uses both density dependence and density-independent stochastic error as a tool to explore harvest strategies for biologists. The inclusion of weak density dependence changes expected harvest rates little from density-independent models. However, it does lower the probability of overharvest at low densities. Alternatives to proportional harvesting are explored to reduce the uncertainty in annual harvests.

Name thesis:degree_name

Ph. D.

Level thesis:degree_level

doctoral

Discipline thesis:degree_discipline

Fisheries and Wildlife Sciences

Department dc:contributor.department

Fisheries and Wildlife Sciences

Grantor dc:publisher

Virginia Tech

Year dc:date.issued

2006

Author dc:creator

• McGhee, Jay D.

Chair dc:contributor.committeechair

• Berkson, James M.

Committee members dc:contributor.committeemember

• Vaughan, Michael R.
• Stauffer, Dean F.
• Brewster, Carlyle C.
• Kelly, Marcella J.

• wild turkey
• theta-Ricker
• Meleagris gallopavo silvestris
• non-linear regression
• harvest
• density dependence

Statement dc:rights

• In Copyright

Licence dc:rights.uri

• http://rightsstatements.org/vocab/InC/1.0/

Dc Identifier Other

etd-02102006-161134

OAI identifier oai:identifier

oai:vtechworks.lib.vt.edu:10919/26164