I am working with Dr. John Skalski at the Columbia Basin Research Center on using likelihood-based methods to reconstruct wildlife populations using primarly age-at-harvest data. I am extending the existing modeling framework to allow for variance components analysis in the demographic processes of harvest and survival. This will permit a parsimonious model structure that accomodates process error in a reasonable way to allow for reduced-parameter modeling, as well as estimation of the effects of environmental stochasticity on population status and viability. I am also extending the models to include the reproductive process, and creating conditional-likelihood models with a second-stage abundance estimate for situations where no auxiliary data are available/ I also intend to assess the ability of these types of models to predict future abundance.
Population reconstructions obtained from this work permit a great variety of population assessments that are typically pieced together from disparate data sources. Assessments can be made of harvest rates, natural mortality, population abundance and recruitment from a single (aged) dataset containing consecutive years of harvest.
Much of my work involves simulation modeling, where I perform model fitting using ADMB, and use R for data simulation and post-fit model comparisons (summaries, plots, tables, etc.)
For now, here are some links to previous work in this field:
- Statistical Models for Population Reconstruction Using Age-at-Harvest Data. Nancy E. Gove, John R. Skalski, Peter Zager and Richard L. Townsend. The Journal of Wildlife Management, Vol. 66, No. 2 (Apr., 2002), pp. 310-320.
- Calibrating Statistical Population Reconstruction Models Using Catch-Effort and Index Data. John R. Skalski, Richard L. Townsend, Brian A. Gilbert. Journal of Wildlife Management, vol. 71, No. 4 (Jun., 2007), pp. 1309-1316.
- Wildlife Demography: Analysis of Sex, Age, and Count Data. Skalski, J.R., Ryding, K.E., and Millspaugh, J.J. 2005. Academic Press, San Diego, California, USA.